You are here: Home > Analyse > Internet Measurements > RIPE Atlas

Technical Updates

Release of probe firmware 4960/4965
  • Fix local time management to deal with cases where the probe did not adjust its local time correctly
  • Add support for reinstalling the current firmware if (filesystem) corruption is suspected
Release of probe firmware 4950 (Anchors only)
  • Fixed a bug in traceroute where a trace was supposed to stop after 5 non-responding hops, but it stopped earlier if the starting hop was not set to 1
  • Added support for DNS cookies (API/UI support is in the works)
  • Added support to set EDNS version, flags and options (API/UI support TBD)
  • Added support to send DNS queries with IPv6 destination option (API/UI support is in the works)
  • Code shuffles to record timestamps for response time as close as possible to sending and receiving calls
New UI release (T'Lara)
  • Major changes to the measurement details pages
  • Update the DNS root instances page to reflect recent changes in the l-root naming scheme
New UI release (Sisko)
  • Restore NSID and SOA functions on DNS measurement maps
  • Minor tweaks to DNS root instances maps
New UI release (Robau)
  • Fix 'type' and 'af' filters (af=4, af=6 and af=null now work as expected)
  • Mobile CSS tweaks for measurement details
  • Enabled a new measurement UI backend for beta testing
New UI release (Quinn)
  • Enable Atlas anchor VM applications and add various support procedures for this
New UI release (O'Neill, Picard)
  • Preparations for a new measurement detail UI release
New UI release (Nelson)
  • Add PATCH of measurement is_public and description, and add tag support
  • Measurement API HTTP host/port/query_string fixes
  • Don't expect the HTTP port to be 80 by default, it varies by scheme
  • Fix measurement option queries in DomainMON
New UI release (Mitsuya)
  • Fixed an issue where probes coverage map was too picky for AS filtering
  • Tweaked cache times on some measurement API endpoints to mitigate some heavy hitting users
Release of probe firmware 4940
  • Add support for DNS client subnet option
  • Add support for HTTP host header
  • Add support for ping synchronous DNS resolution (only used internally)
  • Update SSL/TLS list of ciphers (taken from wget 1.19.1)
  • Add support for SSL/TLS elliptic curves
  • Add better error reporting for TLS errors
  • Add support for traceroute ToS
  • Internal code changes to align the behaviour/scripts used on v3 and v4 probes
  • Fix a bug where probes no longer sent DNS "sos" queries
New UI release (Landry)
  • Improvements to multiple functions such as sponsors, statistics, emailing, ambassadors and credit accounting
  • Improvements to controller handling
  • Added more instances to DNS root server visualisations
  • Improved a case where measurement end times were not properly displayed
  • Fix fallback for permissions without a human readable name
New UI release (Kirk)
  • Fixes to QuickLook functionality
  • Lots of internal restructuring, modularisation of the infrastructure
Release of probe firmware 4930
  • This firmware is a clone of 4910, aimed exclusively for the upcoming v4 probes
New UI release (Janeway)
  • Breaking change: changing HTTP Response from plain "list" to "dict"
  • Improved suggestions for previous measurement IDs in the measurement creation form
New UI release (Ivanova)
  • On measurement creation: reporting back on "Partially Accepted". That means part of the request was valid and we can carry on scheduling measurements, but part was invalid and we report back in the "errors" response field.
New UI release (Hammond)
  • Various internal changes
  • Start removing abandoned probes from anchoring measurements
  • Also exclude decommissioned anchors when filtering for "disconnected"
New UI release (Georgiou)
  • Various improvements for internal scalability
  • Add a 'when' field to the measurement 'status' object
  • One-off measurements will no longer have an 'interval' set
  • Fixed a bug where the download date-range picker didn't always work
  • Enabled measurement tags in various measurement pages (listing, details, new measurement, ...)
New UI release (Felna)
  • Bugfix: filtering for Specified measurements wasn't properly working
  • Bugfix: in some cases measurements disappeared from the list
  • Small fixes for measurement tag adding
  • Enable measurement tag endpoints in API reference
New UI release (Emerson)
  • Add support for measurement tags to result fetching and measurement stopping
  • Add documentation for measurement tagging
  • Fix include=probes not working in participation request API call
  • Add add a few instance IDs to m-root and k-root
  • Use dateRangePicker for selecting date ranges in the measurement results downloads
  • Add page numbers in credit history
  • Fixed an issue in DNSMON where changes to the monitored server set were not visible in the UI
New UI release (Douglas)
  • Use different access logic to non-public probe's non-public builtins, added "Get non-public measurement results" api key
  • Simplify built-in documentation tables
New UI release (Chapman)
  • Various internal changes to stability and scalability
  • Embedding an introductory video to about pages
New UI release (Barnett)
  • Added a list display for all currently disconnected anchors
  • TraceMON is no longer in beta status
  • Fixed showing wrong URLs after measurement creation in UI
  • Fixed wrong axes in credits chart
  • Fix bandwidth limit display on probe status pages
  • Fix documentation for probe archive query filters
New UI release (Ziusudra)
  • Minor internal fixes
New UI release (Adama)
  • Minor changes to application / ambassador supporting procedures
New UI release (Yamato)
  • Added preparations for handling VM anchors
  • Preparing for changes in accounting transaction and probe archive data storage
  • Don't show decommissioned anchors on the anchors map
Release of probe firmware 4910/4915
  • Upgrade busybox (used inside the probe firmware)
  • When name resolution results in a 'local' address and the probe denies executing the measurement, report the offending address
  • Report how long name resolution takes (except for DNS measurements)
  • SSL/TLS certificate measurements now report the cipher that is chosen by the server
  • Fix bug where a TCP socket was not set to non-blocking when explicit binding to an interface was required
New UI release (UNSC Infinity, Voyager, White Star, X-301)
  • Various small bug fixes
New UI release (Spaceball One, Talyn)
  • Add API endpoints for claiming member credits and listing eligibility of associated members
  • Add the new OpenIPMap service to the detail page of traceroute measurements
  • Fix behaviour of filter buttons on the probe coverage map
  • Fix result downloads where the "stop" may have been ignored
New UI release (Quietly Confident, Rotarran)
  • Lots of changes to internal tasks
  • Minor changes and fixes to DNSMON
  • Added supporting calls for OpenIPMap
New UI release (Prometheus)
  • Proper bookkeeping of anchoring measurements when an anchor is decommissioned
  • Document results JSON as of firmware 4750
  • Various fixes and improvements on the measurement listing and form interaction
  • Various enhances to ambassador notifications
New UI release (Olympic Carrier)
  • Various tweaks on UI and API caching
  • Handle edge case for certain API exceptions
  • Don't attempt LatencyMON viz/routequake API for unsupported measurement types + improve LatencyMON performance for built-ins
  • Internal reshuffling of periodic tasks
New UI release (Nebuchadnezzar)
New UI release (Liberator, Moya)
  • Show current UTC time and UTC labels to make it clear to the user that they should be working in UTC
  • Various improvements to probe diagnostics
  • Various changes to anchor provisioning
  • Further work on anchor mesh split
New UI release (Korolev)
  • Added support for "include probe ID" feature in ping measurement
  • Added support for "specify SNI option" in SSL/TLS measurements
  • Added support for setting the timeout in DNS measurements
  • Added support for DNS "macros" feature
  • Document that traceroute represents unrecognised ICMP error codes as integers
  • Improved reference docs for measurements/{pk}/latest
New UI release (Inchon Velle, Jupiter 2)
  • Fixed an inconsistency where API key usage interfered with 404 URLs turning them into 403s
  • Fixed measurement cost gauge behaviour
  • Fixed a bug in the daily cost reporting on some measurement detail pages
  • Split the "anchoring measurements" into "anchor mesh" and "probe anchoring" and update all relevant pages accordingly
New UI release (Heart of Gold)
  • Significant update on the web framework used behind the scenes for RIPE Atlas
  • Fixed a corner case in SSL/TLS parsing
  • Changed the IPv6 target address for B-root, one of the root DNS servers
New UI release (F-302, Galactica)
  • The measurement creation form has been reworked to make use of the API when submitting a new measurement request. This should also improve the user experience when working with this form.
  • Improved the error reporting in the API to make it easier to find out the cause of errors. See the documentation for more details.
  • The "probes" tab on the SSL/TLS certificate measurement result pages has been reworked to show more precise results about common name (CN) / subject alternative name (SAN) name checks as well as other minor additions.
Release of probe firmware 4790 (v1/v2 only)
  • This is a v1/v2 only release meant to fix stability issues that resulted in stretching the hardware capabilities too far.
Release of probe firmware 4780/4785
  • Various fixes for stability and security
  • Allow DNS "macros" like $p, $t, $r, which are replaced by the probe each time with probe ID, timestamp and a random value, respectively
  • Add SNI support to TLS/SSL checks
  • Add option to embed probe ID in ping requests in order to filter out RIPE Atlas generated ICMPs
  • Report time-to-complete in TLS/SSL results even if fetching the actual certificate fails
  • Fixed a bug where there were no results delivered for DNS measurements with the "include the abuf" option turned off
New UI release (Enterprise)
  • Update raw data struct docs to mention a bug in the code and clarify a type.
  • Added better certificate validation for wifi measurements
  • Some updates to the DNS root instance maps
  • Add a feature to send periodic reports to ambassadors
New UI release (Defiant)
  • Cleanups in the API documentations
  • Tweaks for the DNS root-server maps
  • A new tool called TraceMON is available now to visualise traceroute results
  • Minor fixes to some corner cases related to the recent internal changes
New UI release (Black Star, Cloud 9)
  • Preliminary work on wifi opt-in feature
  • Major internal restructuring of the measurement API code
  • Updates to the DNS root-instances visualisation
  • Improve errors provided when asking for probes in invalid prefix/ASN and for DNS query argument validation
Release of probe firmware 4770/4775
  • Fixed a regression where DNS measurements using TCP transport always failed
  • Fixed a bug where WiFi measurements could sometimes report using the wrong outgoing network interface
New UI release (Yggdrasill, Zuberberg, Apollo)
  • Enabled traceroute visualisations for a test group
  • Added internal tasks for wifi opt-in
  • Preparations for probe diagnostics pages
  • Added a separate counter for connected anchors
  • The default behaviour on the coverage map is now to hide abandoned probes (all categories can be turned on/off)
  • TCP traceroute now accepts packet size of 0
  • Fixed an issue where anchor status pages would not load in case the system is running in "reduced data mode" (i.e. during maintenance periods)
New UI release (Unichapel, Vera, Wabbajack, XV)
  • Minor fixes with probe country code notifications
  • Preparations for IPv4/IPv6 stability system tags
  • Internal application server changes
  • Updates to sponsorship terms&conditions
  • Added more DNS root measurements (30001, 30002)
  • Fix deprecated endTime in streaming API
Release of probe firmware 4760/4765
  • Changes to use USB sticks (in v3 probes) as little as possible by buffering results in memory. As a tradeoff for more expected stability, the probes lose recent, unreported measurement results (ie. the ones collected in the last 60-90 seconds).
  • Unexpectedly terminated TCP connections caused SIGPIPE signals crashing the measurement the process
  • Various stability and security fixes to the measurement code
New UI release (Slayer Scythe, TARDIS)
  • Respond with 404 instead of 200 + error-in-JSON for the deprecated v1 APIs
  • Minor enhancements to anchors and ambassadors
  • Changes for decommissioned anchor list and various changes to the decommissioning procedure
  • Fix a bug where country of the probe did not reflect its geo-coordinates
  • Fix an issue where timetravel was not working when trying to parse 7000 range built-ins (connection and uptime events)
Release of probe firmware 4750/4755
  • First release of wifi firmware (4755)
  • Various security/stability fixes in the measurement code
New UI release (Rosetta Stone)
  • Important: the v1 APIs have been decommissioned. v2 is available for all use cases.
  • Preparation work on an upcoming "probe events" page
  • Many other internal changes
New UI release (Pick of Destiny, Quark)
  • Internal changes for stability and scalability
  • Fix a bug that caused the list of prominent probes to be calculated incorrectly
New UI release (One Ring)
  • Enhanced piecharts for current income/expenses
  • Improved manual location picker for probes
  • Bugfix: ordering of in-Atlas messages was not always deterministic
  • More features added to the credits API
  • Better estimates of current and projected credits use
  • Improved websocket-client handling related to ripe.atlas.cousteau
New UI release (Nimbus 2000)
  • Added a new pie chart about current credit status under "my Atlas"
  • Fixed an issue in the measurement specification form where the date selector could pick the year 1899, which is before RIPE Atlas epoch
  • Released LatencyMON ver
  • Added API v2.2.0 with a new credits API
New UI release (Mjölnir)
  • Lots of internal changes for stability and scalability
  • Make the ordering of "my measurements" consistent across pages
  • Only suggest RIPE Atlas anchors when scheduling HTTP measurements
  • Improve consistency in probe notification methods
  • LatencyMON: document showMinimumByDefault
  • Fix a few intermittent JavaScript errors on various pages
  • The measurement creation UI is clearer about it using UTC times
  • Make it easier for users to embed DomainMON in their own pages
New UI release (λαβύρινθος)
  • Numerous internal changes for stability and scalability
  • Added a new "credits API"
  • Fixed a bug that affected the probe view of some SSLcert measurements
  • Added g-root IPv6 to the built-in measurements
New UI release (Jubba Cloak)
  • Optimisations to the caching mechanisms of API output
  • Many updates to the API reference documentation
New UI release (Holy Grail, ICE)
  • Make the measurement listing pages more responsive
  • Some more adjustments to the new API keys structure
  • Make the probe status page more responsive for probes that send excessive SOS messages
  • Fixed permissions for adding measurements into measurement groups
New UI release (Fulcrum, Gungnir)
  • Fix anchor map 'bubble' for IPv4 only anchors
  • Fixes to LatencyMON labels
  • Changed the API keys management: instead of one API key per function, from now on API keys can be assigned permissions, allowing users to combine as many functions to their keys as possible
  • Lots of updates to the root DNS instances maps
New UI release (Excalibur)
  • Lots of internal changes in preparation for an upcoming application framework update
  • Deprecated the v1 API documentation
  • Changes to the measurement detail page layout
  • Changes to page caching, especially on the documentation pages
New UI release (Dyrnwyn)
  • New LatencyMON release
  • Separate out the system-ipv6-doesnt-work explanation for the IPV6 ULA case
  • When probe is connected but there are no SOS messages, show info instead of warning as nothing is actually wrong
  • More robust error checking on participation-requests
  • Fixed issue adding servers to DomainMON zone
New UI release (Crocea Mors)
  • Fixed a typo (tsla_record vs. tlsa_record) in the API
  • Enhance prepend_probe_id field docs
  • Roadmap layout improvements
  • Ambassador management improvements
  • Check ASN entered when applying for an anchor
New UI release (Babelfish)
New UI release (Ancile)
Release of probe firmware 4740 and kernel 1080
  • Disallow measurements towards
  • Internal changes to support wifi measurements
  • Report local network configuration every hour
  • Switch to ext4 filesystem for USB storage
  • Check for read-only USB and send SOS message if that's the case
  • Fixed kernel to not download new firmware if the usb stick is read-only
New UI release (Zatanna)
  • Adding logic to deal with potential IPv4 (or IPv6) only anchors
  • Changes to probe archive API
  • Internal changes for improved stability and security
  • Documenting the new "bonus for results delivered" feature
New UI release (Yorick)
  • Fix adding measurements to an existing DomainMON domain
  • Work on the upcoming probe deployment API
New UI release (Wonder Woman, Xavier)
  • Lots of internal changes about scalability and testability
  • Added total_uptime field to API
  • Fixed a bug where in rare case the scheduler did not start a measurement
  • Fixed some inconsistencies regarding the example API call section in the measurement form
  • Users logins now preserve the full URI of the current page so that (for example) upon return the user is taken to the same tab as before
  • Fixed an issue where one-off measurements initiated in the UI form were rejected
  • Several changes needed because of anchor lifecycle replacements
  • Expose first_connected and total_uptime values to the probe archive API
New UI release (Valkyrie)
  • Various enhancements for API internals
  • Development of the upcoming WiFi measurements
  • Various fixes to DomainMON and probe system tags
New UI release (Superman, Thor, Ultron)
  • Various changes to anchor management functions and the data streaming service
  • Preparations for WiFi measurements (UI, API, probe firmware, ...)
  • Fix some probes with wrong geolocation
  • Fixes to minor QuickLook and API filtering bugs
  • Fixed a bug where maps didn't properly filter for countries
New UI release (Ramona Flowers)
  • As usual: many stability and scalability improvements
  • Added "topology measurements" to all probes as new "built-in measurements"
  • The number of probes was wrong in some cases on the measurement detail page
  • Fixed an issue where some non-DNS measurements could show an invalid include_abuf configuration option
  • Fixed a bug where some probe registrations did not behave correctly
  • Internal preparations for WiFi measurements
New UI release (Quicksilver)
  • Many internal changes for stability and scalability
  • Upgraded bits and pieces of the documentation
  • Changed layout of probes involved in a measurement on the measurement status page
  • Fixed an issue where some traceroutes showed negative packet loss
  • Updated LatencyMON documentation
  • Various minor changes to LatencyMON
New UI release (Poison Ivy)
  • Various enhancements to internal functioning and consistency.
  • Added a new tab to the probe status page to describe various good (and bad) probe conditions
  • Added a "prevent_resolving" field to skip the preliminary DNS check of targets, in case the target doesn't (yet) exist in DNS at the time of scheduling the measurement
New UI release (Oracle)
  • Started to collect historical changes of RIPE Atlas anchors (expecting the first lifecycle events coming up soon)
  • Added an ability to have news items on the opening page
  • Various changes to LatencyMON, deprecate seismograph
  • Anchor-measurements API now exposes type of measurement by default
  • Lots of changes to the API v2 documentation
  • Restarting the "you're out of credits" warning system
New UI release (Northstar)
  • Stability, usability and performance fixes in the UI
  • Changes to l-root IPv6 address
  • Fix an issue in the measurement form not handling reloads of stop time properly
New UI release (Lois Lane)
  • Various internal upgrades
  • Enhancements to the ambassador and anchor administration
  • LatencyMON: improve documentation about the axis
New UI release (Kitty Pryde)
  • Finalisation / documentation of the v2 APIs
  • Release of the "bill to someone else" feature, as part of "sharing credits with colleagues"
  • Stop time in calendar in measurement creation page was confused with UTC
  • Various bugfixes to ambassador checks
  • Various adjustments to users with negative credit balance
New UI release (Jean Grey)
  • Preparations for major application framework version change
  • Various changes in UI to support client-side data table filtering
  • Enabled the "bill to another user" feature in the measurement form and the corresponding credit-sharing feature
  • Fixed a bug in the probe API preventing filtering by distance
New UI release (Iron Man)
  • Added a packet loss column for the probes' list for ping measurements
  • Various changes in UI to support client-side data table filtering
  • Enabled easier navigation from the anchors list to individual anchors
  • Fix inconsistencies in probe bandwidth limitations
  • Added TLSA queries to DNS measurements
New UI release (Harley Quinn)
  • Fix an issue of not using proper ISO8601 times on the measurement probe list
  • Fixed a regression that prevented usage of tags when adding more probes to a measurement
  • Added a better explanation of the "measurement spread" control
  • Enabled support for "my measurements" with API keys in API v2
  • Fixed an issue where data download queries resulted in 500 errors, if the queried time interval did not overlap with the measurement run interval
  • Improved response time when querying for traceroute results for measurements involving lots of probes
  • Allow stopping measurements that have been scheduled to start in the future
  • Fixed an issue which in some cases prevented creating new domainmon zones
  • Revamped the roadmap pages
New UI release (Groot)
  • Major reshuffling of the web site menu structure
  • Many layout changes and enhancements to user interaction
  • Release of a new "overall daily resource use" limit feature, meant to be more flexible that the current max measurements / max probes
New UI release (Elektra, Flash)
  • Introduced "standing credit orders" to regularly transfer credits
  • Fixed a bug in the measurement form about the interaction between area/type fields
  • Fixed a bug about tzinfo on [first|last]_connected in probe status (API)
  • Fix inconsistency about probe bandwidth in the UI
  • Many internal changes in API handling
Release of probe firmware 4730
  • Better error handling for unexpected measurement results
  • Probes will soon start reporting their uptime in a new "virtual measurement", ID 7001
  • Fixed an error case where results of one-off traceroutes could have interfered with ongoing traceroutes
  • Added preliminary support for specifying a timeout parameter for DNS measurements
New UI release (Deadpool)
  • Preparations for credit "standing orders"
  • More maps support filtering on result keys
  • Enhanced the layout for the probe involvement log of a measurement
  • The measurement API is now a bit more forgiving when asking for probes that are not available at the time of specifying a measurement
New UI release (Coulson)
  • Fixes to stability, scalability and the upcoming v2 APIs
  • Fixed a bug where probe SOS messages were not recorded with certain types of probe resolvers
  • Added an experimental feature to limit measurements based on results per day requested, instead of number of probes or measurements
  • The "doughnut" in the credit history page now leads to the measurements mentioned therein
  • The measurement API now denies measurement specifications that are known to be unfulfillable
  • Fixed a bug where the quicklook page didn't update the map and the list after first answer
  • Added anchor IPv4/IPv6 comparison links to more (logical) places
  • Most maps now include the timestamp of the shown measurement
  • Added a "success indicator" to the probe tab for all traceroute measurements
  • Added a quick AS name lookup feature to the list of probes involved in a measurement
  • Added experimental first_connected and last_connected fields to the probe API
  • DomainMON now uses the NSID bit too
  • Added a "RIPE NCC watermark" to some of the visualisations
New UI release (Batman)
  • Fix a case where the number of reported number of currently responding probes could fluctuate as a function of time
  • Put watermarks on LatencyMON widgets
  • Minor fixes to DomainMON
New UI release (Archangel)
  • Various usability fixes
  • Added an IPv4/IPv6 comparison page to all anchors, using LatencyMON. See this example
  • Various fixes in the simple traceroute UI layout (add DNS names, fix layout with late packets)
  • Added HTTP to the list of anchoring measurements
  • Add DomainMON to the menu
New UI release (Yuna, Zelda)
  • Major internal changes due to web framework upgrade
  • Older roadmap entries are now available in the roadmap archive
  • Various usability fixes in measurement details, maps, LatencyMON and DomainMON
New UI release (Vaan, Wario, Xzar)
  • The UDM "region map" is now part of the documentation again
  • The measurement scheduler now recognises the existence of blacklists ("do not probe") and whitelists ("no restrictions on measuring"). You can let us know via email (to if you'd like to whitelist/blacklist your prefixes.
  • Various changes to the upcoming v2 APIs
  • Enable DNSMON tab for all DNS UDMs
  • Added an anchor quick-search feature to the "select probe IDs manually" form
New UI release (Umaro)
  • Minor fixes on DNS root instance maps
  • Fix a bug where probe tags were not suggested properly in the new measurement form
  • Expose built-in measurements in a separate tab in the measurement listing
  • Added more info to the measurement UI about the specification of the measurement
  • Credit archive download now supports fragmented JSON format
  • Fix an edge case where the traceroutes did time travel before history began
  • Various fixes to LatencyMON and its documentation
Release of probe firmware 4720
  • The method for measuring times (e.g. RTTs) for each measurement has been switched to use a strictly monotonic, relative clock, thereby avoiding the artefacts caused by absolute clock changes due to time synchronisation.
  • The RDATA field of a DNS measurement result (in response to a TXT query) is now a list of strings. It was a single string before.
  • The cipher list supported by SSL/TLS certificate checks have been refreshed.
  • Fixed a bug where one-off results were reported multiple times in some cases.
  • Fixed a issue where NTP measurements could generate syntactically incorrect results which, as a consequence, were never stored.
  • Fixed an issue where IPv6-only probes did not properly report their network configuration.
  • Fixed a bug where failed DNS measurements in some cases did not report the time of measurement.
  • The probes, in addition to the infrastructure, now also enforce the restriction that local (RFC1918 and link-local) addresses should not be measured.
  • When removing static configuration from a probe, the statically added previous DNS server was still used
New UI release (Tifa Lockhart)
  • Preparations for "credit vouchers"
  • Changes to probe application procedures related to size of ASN of the host's network
  • Include RTTs and packet loss values in various map popups
  • Various minor bugfixes in time travel maps
  • Fixed a bug where prefix searches did not supply "more specific" results in probe and measurement APIs
  • Added a measurement interval "spread control" feature to the measurement forms
  • Lots of internal changes to keep up with version changes of the underlying web framework
New UI release (Samus, Ratchet, Princess Peach, Q*Bert)
  • Various fixes to LatencyMON
  • Document LatencyMON and expose it to more users
  • Update Sagan to 1.1.0
  • While doing time travel, show 100% packet loss with a distinct colour
  • Traceroute output has been revamped in various places throughout the UI
  • Various changes to ambassador pages
  • Quicklook has been switched to use data streaming for fetching results
  • Fixed a bug in the NTP maps where the time offset could be displayed wrong
  • More additions to the v2 APIs
  • Fixed a bug in the measurement archiver where the end time of measurements was recorded incorrectly
New UI release (Odahviing)
  • Probe uptime log entries are posted real-time now, there's no more delay in the UI
  • It is now possible to assign other individuals as co-hosts for a probe
  • Probes that are co-hosted by me ("shared with me") now show up on a separate tab in the probe list
  • Numerous internal changes for stability and scaleability
  • Fixed an issue map filtering only worked once per page load
New UI release (Navi)
  • Added support for packet-interval for ping measurements
  • Probe API now allows filtering available probes by address family (af)
  • Upgrade ripe.atlas.sagan to 1.0.0
  • Show total number of probes on available maps
  • Fix for case where probes appeared twice in a probe API filter
  • Fix for grey markers appearing when filters are enabled on some maps
  • Fixed a visualisation bug in the credits history view where not all charts showed up
New UI release (Mario)
  • Added an option to the measurement API to filter the list to "my measurements"
  • Added a "time travel" feature to virtually all maps of ongoing measurements to check the status of the map at a certain point in the past
  • DNS maps for UDMs now have multiple variations of result visualisations (RTT, SOA or answer)
  • Enabled HTTP measurements towards RIPE Atlas anchors
New UI release (Lara Croft)
  • Major enhancements to visualisations of UDM measurements (see the maps for these measurements)
  • Show IPv4/IPv6 properties for UDMs consistently all across the site
  • Added "time travel" feature to UDM maps / probe listings
  • Lots of preparations for upcoming visualisations
New UI release (Kirby)
  • Fixed a corner case where measurements with no data plotted maps with all the probes and no colour.
  • Started enforcing a pagination limit in UI and API for probes and measurements
  • Do not show the seismograph for one-off pings
  • Show the network details of public probes in UI
  • Enabling gzip compression for the measurement API data downloads
  • Many internal changes mostly to prepare the way for new visualisations
New UI release (Jim Raynor)
  • Lots of internal changes as usual
  • Small enhancements to the section showing the probe's public IP address
  • Fixed a bug with possible redirection loops after login
  • Added a pagination limit to the measurement and probe APIs
  • Minor fixes with result maps not always showing correctly
New UI release (Herman Toothrot)
  • Many internal changes for stability and scalability
  • Maps for DNS measurements have been extended to visualise RTT, answer and NSID diversity
  • Fixed some minor issues in the "probe down" notification settings
  • Fixed an issue which in some cases prevented DELETE/stop measurement from working (since GLaDOS)
  • NTP measurements now also have a useful map visualisation
Release of probe firmware 4700
  • This is mostly a maintenance release, with internal behavioural changes only.
New UI release (GLaDOS)
  • Major internal changes to the measurement API (not visible to users)
  • Fixed a bug where the ASN/country filter did not work on the network coverage map
  • Make the "built-in measurements" tab of the public probes' status page visible for all users
  • Minor tweaks to the anchor list and the anchor-as-a-probe page
  • Change the colour scheme used for the maps to a more distinctive set
New UI release (Fox McCloud)
  • Various internal changes
  • Added some new k-root anycast instance names
  • Added support for CD bit in DNS measurements
New UI release (Ezio)
  • Various internal changes
  • Added region name suggestions when adding new measurements using manual probe selection
  • Fixed an issue when "LIR IPv6 connectivity test" failed in certain cases
New UI release (Donkey Kong)
  • Minor tweaks to the cost estimator in the measurement creation page
  • Fixed a couple of bugs in the new maps
  • Many internal optimisations to make the UI faster
  • Applied minor changes to the global page template
  • Documented the existence of system-ula and system-rfc1918 probe tags
New UI release (Cloud Strife)
  • Added a table to the probe status page to show the results of the various probe IP discovery methods
  • Lots of small fixes and improvements to the maps
  • Added a human friendly view of traceroute results
New UI release (Bowser)
  • Deployed a whole new version for displaying all maps; now it includes hiding/showing of some results, better filtering and more
  • Fix for some cases where a statically specified DNS server would not be shown as such in the probe network config page
  • The "LIR IPv6 traces" function has been renewed; it includes IPv4 traces too
  • Added a page to documentation about built-in measurements
  • Improve documentation on result streaming
UI changes (Aerith Gainsborough)
  • Change the page layout of the probe's built-in measurement tab to be responsive (i.e. behave better when the browser window is narrow)
  • NTP measurements have been added to the measurement UI
  • Many internal changes
UI changes (Zahl)
  • Major internal changes to accounting (credits) with minor UI related changes
  • Fixed a bug in probe selection by list of probe IDs
  • More alignment to the new house style
  • API call to get metadata of a measurement now appears in the measurement info page
UI changes (Yridian)
  • Major layout changes: switched to the new RIPE NCC UI design/layout
  • Fixed an issue on the probe deployment map where disconnected probes may have shown up as connected
  • Ambassadors can now mark their attendance at various conferences
  • Display the ASN name(s) in the probe status page
  • Fixed an issue where "today" was not available as a potential end time when downloading results
  • The "written off" status is now displayed in the probe status page, giving an indication to the viewer about potential issues
  • Fix dropdown and datepickers in API key generation page
  • Remove the 'port' field for non-TCP traceroutes
  • Disable the Target field for use-probe-resolver
Release of probe firmware 4680
  • This firmware incorporates a few bug fixes:
  • If probe has statically configured nameserver and also DHCPv4, the DHCP one wins, thereby fixing stale DNS entries
  • Enhance the NTP client on v3 probes
  • V3 probes now pick up IPv6 DNS resolvers from RA messages
  • Stability issues on v1/v2 probes when HTTP measurements immediately fail with connection errors
  • "SSLCert" measurements now also support TLS
New UI features and enhancements (Vulcan)
New UI features and enhancements (Son'a)
  • Many internal changes
  • It is now possible to select probes in the measurement specifications UI without using the interactive map ("new set - manual" in the probe selection section)
  • For ordinary users show IP and controller in list of connects
  • Be more verbose in warning on automatic geolocation or no geolocation at all, and asking the user for a real location
New UI features and enhancements (Romulan)
  • More UI conversion for the upcoming layout changes
  • Added a feature to search for probes by "centre" and "distance" as filter
  • UDM UI documentation has been updated to match the new UI
  • Add the number of probes involved in a measurement to the measurement listing page
  • Enhanced the probe auto-geolocation feature and relevant UI bits
New UI features (Q)
  • Some fine tuning on the new measurement UI
  • Many internal improvements
  • Added a new banner to notify users about the upcoming new Terms & Conditions
  • Many pages have been re-stylised in preparation for a new visual layout
  • Allow searching in the probe API by using a geographical "bounding box"
New UI features (Prophet)
  • The old measurement UI is now officially deprecated, the revamped new UI/API is now the default
  • Many pages have been re-stylised in preparation for a new visual layout
Release of probe firmware 4670
  • Fixed two bugs in ping measurements where the probe had issues pinging its own IPv6 address
  • Fixed wrong host header for IPv6 literals in http measurements
  • Probes are now trying to avoid starting too many measurements at exactly the same second
Cost estimation corrected
  • We've finally fixed the lingering "measurement cost estimator is wrong" bug
New UI features and enhancements (Monean)
  • Many user-facing enhancements in the UI:
  • Probe connection status is available via API as a field
  • Allow filtering on description in the measurement API
  • Revamp the API key management page
  • Tweaks to the measurement UI
  • Allow maps to un-cluster result pins
  • Show relevant RIPE Atlas internal messages to users
  • Introduce separate quotas for starting one-off vs. ongoing measurements
  • Exclude "built-in" measurements from the calculation when checking for maximum parallel measurements against a target
  • Add documentation on streaming API
  • Enhance the coverage map with more countries and country names
  • Allow users to use probe tags when scheduling measurements
  • Tweaks to the probe auto-tagging feature
  • Added link to OpenIPMap for traceroute measurements
  • Community/leaderboard now uses "measurements contributed" instead of probe uptime
  • Display all IPv6 addresses reported by the probes
Release of probe firmware 4660
  • Bugfix: fixed a bug in paris traceroute where the ICMP version would have the wrong paris id in outgoing packets
  • Bugfix: fixed a memory leak in the DNS measurements code
  • Enhancement: include a cookie in outgoing ping packets and check the reply
  • Enhancement: in ping, report IP version and target address even if socket connect fails
  • Experimental suport for NTP measurements (not publicly available yet)
Recent UI changes
  • Recently we added the following UI features:
    • Probe tagging: you can now add various tags to your probe to describe the network situation (NAT/NONAT, connection type, etc.)
    • Revamping of maps: we introduced pin aggregation in order to avoid overloading the users' browser
    • Various enhancements to the network coverage page
    You can find more details here: RIPE Atlas Midsummer Update 2014
  • In addition, DNSMON became a production RIPE Atlas service in July.
Release of probe firmware 4650
  • Ping interval option (-i option, needs support from API and UI)
  • The "lts" field is now also available in the output of traceroute, dns, sslgetcert, httpget
  • Traceroute IPv6 extra error code 'h': destination unreachable/beyond scope
  • Fixed error in the output of traceroute
  • Fixed bug in parsing multiple IPv6 extension headers in traceroute
  • Fixed bug in DNS where RA flag was set in requests
Release of probe firmware 4610
  • The new firmware fixes a few DNS related bugs:
    1. In some cases the probe sent DNS results too often
    2. Querying the local DNS resolver could result in multiple results (one per resolver), but these could not be accessed in the data store. The new version collects results from all resolvers into one data structure. See the data structure documentation page for details.
  • In addition, this firmware adds preliminary support for using IPv6 extension headers. This will be available in the UI and in the API at a later date.
New probe User Interface
  • Today we enabled the new probe User Interface, which has support for:
    • a streamlined layout
    • better search facilities
    • probe tagging (home/office connection type, personal tagging, ...)
    • favourite probes, sponsored probes, abmassadored probes, ...
Release of probe firmware 4600
  • The new firmware release (4600) contains bug fixes for missing fields in ping results (TTL and source address). It also incorporates a more secure way for the probes to authenticate new firmwares before upgrading.
"UDM Status Checks" released
  • Today we released a new feature: UDM status checks. These allow users to check the status of ongoing measurements, and connect this to their own monitoring systems in order to monitor connectivity, raise alarms, etc.
Seismograph updates
  • The seismograph has been updated to handle "raw" ping results, in order to show minimum/average/maximum and packet loss information, in addition to the existing views. The probe selection and the zooming functions have also been enhanced.
Release of probe firmware 4580
  • Due to an issue with the latest firmware release, a subset of the v3 probes were listening to incoming connections on an open port that should not have been left open. As a secondary measure, however, access to this port required credentials only available to the RIPE Atlas probe developers. It therefore never presented open access to the probes. This port (SSH) is used for development purposes in our internal development environment.
  • We upgraded the v3 probes to a new, corrected firmware version (4580), and improved the checks in our firmware release process. The new firmware is otherwise functionally equivalent to the previous one.
  • This issue did not affect version 1/2 probes and anchors.
Release of probe firmware 4570
  • Upgrade kernel on v3 probes without losing static network config
  • Uniform interpretation of the size parameter of ping and traceroute: the size excludes the IPv4/IPv6 header and the transport (ICMP, TCP, UDP) header
  • Add TCP mode to traceroute measurements
  • Most measurements (except for DNS "use probe's local resolvers") now pick up a new list of resolvers if it has been updated by DHCP
  • Support for SRV and NAPTR in DNS measurements
  • Support "number of retries" option in DNS measurements
  • Note that the use of the new measurement flag will be enabled in the UI and API at a later stage.
Various UI enhancements
  • Improvements to maps: a subset of the result maps now allow filtering by country code and ASN
  • Improvements to probe application / registration
  • Making the results of anchoring measurements available: the anchor list now contains links to the measurements associated with particular anchors
New ping graphs
  • Replaced the result graphs for "built-in" ping measurements with zoomable, client-side graphs. This was done in order to keep up with the increasing number of probes.
Seismograph, zoomable RTT graphs and more
  • The Seismograph is a new feature designed to visualise ongoing (ping) measurements. It's sensitive enough to show you otherwise invisible patterns and correlations between various probes.
  • Built-in ping measurements are now visualised based on an interactive, zoomable client-side widget; from now on you can explore your probe's history in full detail!
  • You can now execute measurements more often (i.e. the minimum wait time between measurement iterations can be set lower than before).
  • Downloading of results has a more fine-grained control: you can now download data with intervals of a second's granularity.
  • There's a new map available for RIPE Atlas ambassadors.
  • We've released a new probe firmware: 4550. Changes include:
    • Support for one-off SSLGetCert measurements
    • Reports response time of SSLGetCert measurements
    • ping now reports when the probe's clock was last found to be accurate
    • Change of one of our "registration servers" ("teyla" to "oneill")
    • Various enhancements to HTTP measurement reporting (anticipating that HTTP measurements may become publicly available in the future)
    • Report source address for ping measurement even when no reply is received
    • Small fixes to the DNS measurement and the one-off daemon
    • Fixed bug in UDP/IPv6 traceroute where it would not set the source port correctly
    • v3 probes: makes sure that no DHCP client remains running when a statically configured IPv4 address is used
    • v3 probes: enables compression on the communication channels towards the infrastructure. This should reduce the amount of network traffic needed for reporting results
    • v3 probes: file system check on USB partitions
Major server moves
  • We made changes to a significant portion of our infrastructure in order to increase the availability and stability of RIPE Atlas. This includes moving functionality to new servers, adding load-balancing capability and duplicating some components.
Reporting tools
  • We released some prototype reporting scripts that can be used to analyse results supplied by RIPE Atlas. See more info in RIPE Labs
Quick Look
  • RIPE NCC members can now ask for "quick looks": one-off measurements, bundled with real-time visualiastion of the incoming results, without the need for Atlas credits!
Measurements source code release
One-off measurements, APIs, more visualisations and increased quotas
  • You can now do "one-off measurements": these allow you to ask the system for a single run, quick measurement – you can expect results within seconds!
  • There's now two shiny new RESTful APIs: one for measurements and one for probes. These allow you to do scripted measurements and to get to the results without using the UI manually.
  • We introduced a customised "dashboard" as the opening page, if you're logged in and have some measurements or probes.
  • The "credit reporting" page has been restructured to give you simple visualisations about your current credit status and the ability to see your credit history.
  • Latest ping measurement results are now shown in a neat bar chart layout, as well as a map layout, in order to provide an easy-to-understand overview.
  • We added a new map to show country coverage of Atlas probes.
  • It is now possible to share your probe with your LIR colleagues (provided that they are recognised members of the same LIR and they have visited RIPE Atlas before).
  • We've significantly increased the maximum number of measurements you can run in parallel, and the maximum number of probes you can involve in each measurement.
Probe firmware version 4520
  • This is a bug fix release for all probe architectures. There is a small bug in the probe measurement code that can be trigger only if a controller sends the wrong commands to a probe.
Probe firmware version 4510
  • This release is only for the new v3 probes. It contains fixes for static network configuration.
New layout & new probes
  • We changed the RIPE Atlas website navigation to make it easier to access frequently used functions.
  • When transferring a probe, the system offers suggestions about your colleagues (within the same LIR, if you're a member).
  • You can now change your Single Sign-On email address using the RIPE Access control panel.
  • We added a community page with some interesting "probe locations".
  • From now on, we enforce the credit limits: we send warning emails if your credit balance is approaching zero and we begin to stop measurements if you run out of credits.
  • We introduced the v3 RIPE Atlas probe. The exact model is a TP-Link MR-3020 (3G/WiFi) router, with an additional USB stick, running our custom firmware - so it doesn't act as a router any more, only as a probe.
Probe firmware version 4500
  • RIPE Atlas probe software now supports two more architectures: TP-Link (for the next generation probes) and CentOS (for RIPE Atlas anchors).
  • There is now support for one-off measurements for ping, traceroute, DNS, and HTTPget.
  • We fixed a bug in DNS measurements in which, when querying local resolvers, more queries went to the last resolver.
  • Fixed "error" : { "TUCONNECT" : "Success"}. Before this version DNS TCP and HTTPget reported an error message "Success".
Various UI enhancements
  • Early (beta) versions of the measurement REST API and the probe REST API are available now.
  • The credit model and the known bugs and limitations have better documentation now.
  • It is now possible to transfer a probe to another Atlas user. Keep in mind that if you transfer your probe to someone else, you give away control of the probe to the new host. Also, the probe will generate credits for the new host from then on. The only way to "get back" your probe is for the new host to transfer it back to you.
  • RIPE NCC members now get "suggestions" about their colleagues registered in RIPE Atlas, when transferring probes or credits.
  • RIPE NCC members can now get 1 million credits if they request it.
  • It's now easier for RIPE NCC members to apply for Atlas probes, starting from the LIR portal.
  • We've added a feature about warning users if they are likely run out of credits soon, as well as stopping measurements if users ran out of credits.
  • You can now define an API key to download results of all your measurements, not only a specific one.
  • Atlas anchors are now shown as potential targets when specifying a measurement.
Various UI enhancements
  • You can now download you credit reports for previous months
  • The RIPE NCC Access (single sign-on) subsystem now recognizes RIPE NCC members, so from now on we can develop more features that require recognition of particular members and their registered users.
  • The probes' "Assigned UDMs" section now shows entries about previous measurements, so hosts can look up what the probe did recently.
  • In preparation for the upcoming Atlas anchors and newer probe hardware models, the probe's architecture is now shown in the UI.
  • Auto-notification about down probes is now using a "three strikes" model.
  • Probes can now be earmarked for sponsors, so a sponsor's probe list is now populated with all their sponsored probes, not only the activated ones.
  • We've added a "top hosts" community page.
  • We've added a "top spenders" community page.
  • UDM data downloads are now much more responsive, also with large result sets.
Probe firmware version 4480
  • Fixed bug in traceroute when it has to deal with rfc4884 objects (mpls) that have a wrong size.
  • Delayed DNS name resolution in ping and traceroute. This feature will soon be enabled through the UI.
  • Fixed bug in HTTP GET where some characters where not properly escaped in generating the result JSON.
  • Fixed bugs in the libevent stub resolver to better handle DNS errors and timeouts (affects mostly httpget)
  • Limit the amount of measurement data that is sent as one unit. This prevents probes that have not connected to a controller for some time from overloading the controller.
  • The probe uptime is now in the DNS SOS messages that are sent by probes before they try to connect. This will allow making a distinction between various reasons for disconnects: e.g. probe reboot vs. network problems.
Various UI enhancements
  • This update brings a number of internal changes as well as a few additions to th UI:
  • API keys: similarly to the API keys used in RIPE NCC LIR portal, it's now possible to manage "API keys" in RIPE Atlas. At this time these keys can be used to download and share your measurement results (probe built-in measurements and UDMs). More information can be found in the documentation section.
  • There's a new graph about the credit consumption rate of the last two months; it allows visual inspection of your credit generation/consumption trends. From October on we'll also make your "transaction statements" downloadable for each previous month. You can find both of these by clicking on the "Archives" button at the bottom of the "My Credits" section of the UDMs menu item.
  • Applying for probes: we made simplifications to this procedure which allows easier communication between a would-be-host and us, as well as a more logical procedure flow.
  • We added the possibility to filter the output of the "UDM list". This comes handy for the users who have many UDMs.
  • Various smaller enhancements: flipping through pages in the UDM list should be faster now you'll a separate credit transaction for each of your probes, in case you have more than one
DNS and SSL-certificate measurements
  • We are happy to announce DNS & SSL certificate measurements, a new addition to the UDMs, making it in total 4 types of customized measurements you can perform. In order to make more types of measurements possible, we have also introduced the new web interface: so-called wizard. The old interface has been deprecated, so the usual link now leads to the new wizard.
IPv6 traceroute visualisation for RIPE NCC members
  • We've released a prototype tool to visualise the result of traceroute6 measurements done by all IPv6 capable RIPE Atlas probes. See the article Measuring IPv6 Connectivity with RIPE Atlas - How Do Others See Your IPv6 Network? RIPE Labs article for details.
Probe firmware version 4470
  • This firmware fixes two bugs. The first one is that DNS results may get mixed up when a probe runs two DNS measurements at the same time. The second one is where traceroute sometimes reports a timestamp of 0. In addition, the firmware now has IPv6 literals for the registrations servers, so an IPv6-only probe can connect to a registration server even if it doesn't have a DNS resolver.
Firmware 4470 to be released soon
  • Probe firmware version 4470 will be released soon. This version includes a small data structure change from 4460 which only affects DNS measurements: the addition of the optional field 'src_addr'.
  • We have also published a correction for version 4460, which mistakenly listed a field 'src' for DNS measurements. This field does not exist in 4460.
Probe firmware version 4460
  • The main new feature in this firmware is the use of libevent and rewriting the measurement code to use it. This provides a much higher capacity for doing measurements. In addition, the probe now reports results in JSON. Expect the traceroute output to be completely different. The are a number of small changes. For example, the DHCP client now sets the client-id and vendor class. Please note that if you have given the probe a static IPv4 address through DHCP and the probes stops working then this may be caused by the change in DHCP client-id.
New firmware release
  • A new version of the RIPE Atlas probe firmware will be released later this month. This release will affect the raw data structure of measurement results.
Integration with RIPE NCC Access
  • From now on, the user authentication behind RIPE Atlas is done through RIPE NCC Access, the RIPE NCC single sign-on system. In order to use RIPE Atlas, you'll have to log in with your single sign-on credentials. This change allows us to offer easier access to other RIPE NCC services, as well as to provide more specialised content to RIPE NCC members in the future.
UI changes and a bit more
  • We've deployed some changes to the UI: The most notable change is that from now on the download format of your probe's measurement data is JSON. The actual syntax is pretty much self-explanatory. The DNS anycast map now visualises server versions and SOA serials too. We made further changes to the way we show static network configuration of a probe on the "probe status" page. We're making preparations to include Atlas in RIPE NCC Access, the RIPE NCC single sign-on system. Beyond these changes, we've made some enhancements to the scalability of the upcoming User Defined Measurement feature and did some additional internal ("invisible to the users") modifications.
Probe firmware version 4310
  • This firmware contains a number of small fixes that deal with exceptional conditions. Probes should also reconnected slightly quicker after a disconnect.
New UI features
  • The probe's network usage is now visualised on the probe details page. We show both bits/sec and packets/sec, as seen on the probe's network interface. Note that these include broadcast traffic, so it most likely overestimates the actual network traffic related to the probe. You can now easily look at the last traceroute(6) result of each built-in measurement: click on the link of the "Target Address" column. Downloading the logs for a measurement done by your probe has been moved to the probe details page. The previous method (allowing downloads from the probes grid) did not scale well. A probe host can now ask for notification (via email or an http POST) if the probe is disconnected from our network for more than 30 minutes. You can turn on this feature in the probe settings page. A probe host can also ask for a monthly probe connection report, also in the probe setting page.
1024 probes online
Network coverage maps and details
New firmware released: 4280
  • This new firmware fixes a couple of bugs in the probe firmware, including one that prevents the probes from reconnecting to our network (for up to 3.5 days) in low memory conditions.
Root DNS anycast map available
  • Today we've enabled the DNS root anycast instance map that was announced at the RIPE meeting and in a RIPE Labs article. Clicking on the Maps link in the menu bar (top right) leads you to a page that lists all maps we have available at the moment, including the anycast map. As a small enhancement we also begun to link AS and prefix information to RIPEstat, to facilitate looking up more information about these resources. This is visible on the "probe details" page as well as on various maps (when clicking on individual probes).
UI upgrades
  • The probe configuration page now has a better explanation on which static configuration options were the probes able to use. There's a new global map available: Comparative DNS Root Server RTT map.
Probe firmware version 4270
  • More pings to fixed destinations Traceroutes to fixed destinations DNS root server anycast instance checks
Infrastructure upgrades
  • We're doing major upgrades to the backend infrastructure, in order to prepare the system for more resilience and to the User Defined Measurements functionality.
Probe firmware version 4030
  • We're in the process of rolling out a new firmware version. It enables a new feature that has been asked by members of the community: static network configuration. Using the UI, one can ask the probe to try to use static IPv4/IPv6 addresses (and DNS resolvers). If these settings don't work, the probes will fall back to using DHCP. As a byproduct, this feature also allows IPv6-only deployments.
Notifications about "down" probes
  • We started automatically notifying hosts whose probe seems to be down for some reason. You'll get an email if your probe does not seem to be connected for an extended period of time (the current setting is five days). We hope this helps keeping probes up, especially in "ooops, sorry, I forgot!" situations.
Updated probe geolocation
  • We've introduced some semi-randomness to the geo-coordinates when showing probes on the maps. This solves the issue of some probes not showing up on the map (or more precisely, hiding behind other probes) if there are more than one probes geolocated to the exact same spot.
RTT maps for built-in measurements
  • We've enabled a new feature (look for "Maps" in the main menu) where you can see how each probe sees the RTT to the built-in destinations.
Distribution of new Atlas probes
  • Our Customer Services department started to contact host applicants in order to distribute RIPE Atlas probes. We're contacting those applicants first who score high in our ranking system, i.e. whose network data and geographical location seems to be the most useful for the Atlas network.
Probe DNS and uptime reporting changes
  • We've now enabled two features that were requested by hosts earlier:
    • The "Probe DNS" function allows you to ask for a DNS entry for you probe. Operational details:
      • The exact DNS name of your probe depends on your settings (simple or obfuscated), and can be looked up on the "Probe Conf" section of the probe status page.
      • The zone file is re-generated every 15 minutes.
      • The resulting address is the one we observe when the probe to connects to our infrastructure, so if you're using NAT then it's the public address of the NAT device. Currently it's either IPv4 or IPv6, but not both. We'll add the "other address" too in the future.
    • The probe uptime section has more details. It now contains:
      • Relative uptime for the last week and month, and for the full lifetime.
      • The source IP address of the individual connections.
      • The amount of up- and downtime (in days/hours/minutes), for each connection.
New UI design
  • The overall design of the RIPE Atlas website was upgraded, in sync with the RIPE NCC website redesign.
Probe firmware version 4020
  • The new firmware version enables the use of a second registration server ( All probes are expected to upgrade automatically in the coming days.
New UI features
  • Router type: optional, describes the router/NAT model you have, mainly in a home installation. This information could be useful for us to correlate weird probe behavior across similar router models. Allowed bandwidth: it's not used yet, but you can imagine the future purpose for it :) DNS options: this has been asked by some of you before. By default it's off. If you set it to "simple" then your probe will have a DNS entry. If you set it to "obfuscated" then your probe will have a DNS entry. We'll activate this feature soon. Public: you can make your probe visible to everyone, which means that all Atlas users will see the probe in their probe list (with role "Viewer"), and they will be able to see the measurement details on the probe status page. Consider this an experimental feature -- we have to see if it's useful or not.

More information

There is more information on the FAQs page. If you have a question that's not answered there, please send an email to

We develop RIPE Atlas in cooperation with the Internet community, and we want to know what you think. Find out how to get in touch.