Differences

This shows you the differences between two versions of the page.

--- civilinfrastructureplatform:journal [2018/05/18 13:21]
rajmarshall
+++ civilinfrastructureplatform:journal [2018/08/15 08:11]
rajmarshall
@@ Line 2: / Line 2: @@
 This is the journal followed by CIP developers to keep track of the work being done on behalf of CIP.It starts in 2018Q2.
+===== August =====
+==== Wednesday 2018/08/15 ====
+=== rajm ===
+  * Ran a BBB health check successfully with an email to the list.
+==== Friday 2018/08/10 ====
+=== rajm ===
+  * Created a script for the provisioning of the 2 docker repos and the build script however the network problems within docker are causing the script to exit (not unexpected!)
+  * Tried again to extract builds from the lucj version of kernelci-docker I can wget from within the container but not externally.
+  * Added further points for the issues in [[https://gitlab.com/cip-project/cip-testing/testing/issues/192|#192]]
+  * Ran a BBB health check successfully with an email to the list.
+==== Thursday 2018/08/09 ====
+=== rajm ===
+  * Updated [[https://gitlab.com/cip-project/cip-testing/testing/issues/192|#192]] to list - in the description - the issues discovered so far which will need incorporating into CIP's version of the docker containers.
+  * Started the kernelci-docker to check problems were as before - but also verifying that it does run alongside lava-docker.
+  * Also tried the lava-docker test from the office network - using the B@D builds and still get a failure - having to use the host IP rather than localhost as the port isn't forwarded to the container.
+  * Ran a BBB health check successfully with an email to the list.
+=== toscalix ===
+   * Chair of the weekly IRC meeting
+      * Sent the [[https://lists.cip-project.org/pipermail/cip-dev/2018-August/001470.html|minutes]] to cip-dev mailing list.
+   * Sent a proposal to cip-dev to [[https://lists.cip-project.org/pipermail/cip-dev/2018-August/001471.html|manage the user roles]] in gitlab.com
+==== Wednesday 2018/08/08 ====
+=== rajm ===
+  * Intending to try the BBB health check on lava-docker using our B@D builds - but I still see the UBoot timeout.
+  * I tried provisioning lava-docker without VPN enabled (from outside the company network) and I still had timeout failures.
+  * A fresh provision for week 32 [[https://gitlab.com/cip-project/cip-testing/testing/issues/149|#149]]
+  * Ran a BBB health check successfully with an email to the list.
+=== toscalix ===
+   * Based on the agreement that took place at DedConf 2018, moved the CIP-dev IRC meeting an hour earliues, so people at JST can join.
+==== Friday 2018/08/03 ====
+=== rajm ===
+  * I've successfully built with beaglebone enabled - using connection_command. At the moment it is failing in UBoot though to get here I'm having to open a connection to the BBB from the host and leave it logged in (as we did in the early days of the project) will eventually need a login script adding. The IP address of the host is currently hard coded in boards.yaml.
+  * With a fresh build of the docker image for lava-docker the QEMU health check just works. I wonder whether the network issues I'd been having cause the workers to go irreparably offline? Now looking to build with a beaglebone board configured.
+  * Ran a BBB health check successfully with an email to the list.
+==== Thursday 2018/08/02 ====
+=== rajm ===
+  * 1:1 with toscalix
+  * The DNS failures on needing to restart docker are actually networking failures.
+  * The health check is not running because the worker is marked as offline. I've tried to put it online without success and created another one which also stays offline.
+  * Ran a BBB health check successfully with an email to the list.
+==== Wednesday 2018/08/01 ====
+=== rajm ===
+  * The LAVA docker device is marked as having a health check but it hasn't run, investigating why.
+  * Ran a BBB health check successfully with an email to the list.
+===== July =====
+==== Friday 2018/07/27 ====
+=== rajm ===
+  * I have successfully built lava-docker [[http://github.com/kernelci/lava-docker|from here]] - I had to move the docker directory from /var/lib to give me more space and to restart docker a number of times on getting DNS failures. Also added some `apt-get update`s to the build-lava script.
+  * Ran a BBB health check successfully with an email to the list.
+==== Thursday 2018/07/26 ====
+=== rajm ===
+  * Submitted [[https://github.com/kernelci/kernelci-docker/issues/12|#12]] to kernelci-docker so see if there's a comment, not sure whether it's my settings or how I'm running the interaction.
+  * Submitted [[https://github.com/kernelci/kernelci-docker/issues/11|#11]] to kernelci-docker otherwise start.sh doesn't keep a clean version.
+  * Ran a BBB health check successfully with an email to the list.
+=== toscalix ===
+   * Told Fabrizio from Renesas that the patch review will take a few days since Ben H. is at DebConf.
+   * Ben H. reviewed the CIP Core packages prioritization policy and sent it to cip-dev. I went over it.
+==== Wednesday 2018/07/25 ====
+=== rajm ===
+  * Still getting 500 errors when I start the containers from the original repos and try a build. The build.py file needs modifying on each restart of the containers, this needs scripting? I tried a fresh start of docker (and the laptop) but that still gave the same problem.
+  * Closed [[https://gitlab.com/cip-project/cip-testing/testing/issues/178|#178]] - it should have been done last week when the associated MR was merged.
+  * Sent a status of CIP testing email to the cip-dev list.
+  * A fresh provision for week 30 [[https://gitlab.com/cip-project/cip-testing/testing/issues/149|#149]] - no longer need to add a local test section with last week's merge.
+  * Issue [[https://gitlab.com/cip-project/cip-testing/testing/issues/193|#193]] has been closed by Zoran, it was a problem with his initialisation script.
+  * Ran a BBB health check successfully with an email to the list.
+=== toscalix ===
+   * Sent to Jeff the mails related with the suggestion to contact Mitsubishi Heavy Industry as potential candidate to join CIP.
+   * Catch up with mails from CIP after several week travelling and on vacation.
+==== Monday 2018/07/23 ====
+=== toscalix ===
+   * Attended to the CIP TSC meeting
+      * Reported about bwh and rajm latest work.
+      * Several meetings related with kernel maintenance and Debian LTS are expected during DebConf 2018 in which bhw presence would be needed. No schedule yet.
+      * A mail was sent by Daniel Sangorrin to cip-dev about the criteria to be used to prioritize the package list to be sent to Freexian / Debian LTS.
+==== Friday 2018/07/20 ====
+=== rajm ===
+  * Looking at [[https://gitlab.com/cip-project/cip-testing/testing/issues/192|#192]], managed to get start.sh in the original [[ http://github.com/kernelci/kernelci-docker|repos]] to run to completion - I had to `sudo service docker restart` numerous times to get there - but I'm getting a 500 error on running the kernelci-build/build.py script.
+  * Investigated [[https://gitlab.com/cip-project/cip-testing/testing/issues/190|#190]] and updated with a question to the OP.
+  * Ran a BBB health check successfully with an email to the list.
+==== Thursday 2018/07/19 ====
+=== rajm ===
+  * I've updated Virtualbox to 5.2.16 - looking at [[https://gitlab.com/cip-project/cip-testing/testing/issues/193|#193]] - and after getting the vagrant plugins up to date  a pre-existing VM successfully booted and a fresh provision also worked ok.
+  * Ran a BBB health check successfully with an email to the list.
+==== Wednesday 2018/07/18 ====
+=== rajm ===
+  * Updated Vagrant to 2.1.2 to investigate [[https://gitlab.com/cip-project/cip-testing/testing/issues/193|#193]] I don't see the error, will update to Virtual box 5.2.16 and see if that produces the error.
+  * Merged  [[https://gitlab.com/cip-project/cip-testing/board-at-desk-single-dev/merge_requests/61|MR 61]] - making the final output of create_test.sh more appropriate to the user's situation and closing [[https://gitlab.com/cip-project/cip-testing/testing/issues/178|#178]].
+  * Merged [[https://gitlab.com/cip-project/cip-testing/board-at-desk-single-dev/merge_requests/62|MR 60]] - using a fixed version of LAVA so we can check any new version before incorporating it - and closed the associated issue [[https://gitlab.com/cip-project/cip-testing/testing/issues/191|#185]].
+  * Merged [[https://gitlab.com/cip-project/cip-testing/board-at-desk-single-dev/merge_requests/62|MR 62]] - the moving to a local test - and closed the associated issue [[https://gitlab.com/cip-project/cip-testing/testing/issues/191|#191]].
+  * Ran a BBB health check successfully with an email to the list.
+==== Friday 2018/07/13 ====
+=== rajm ===
+   * Created [[https://gitlab.com/cip-project/cip-testing/board-at-desk-single-dev/merge_requests/62|MR 62]] to use a local test for the BBB health check to make the test more self-sufficient.
+   * Opened [[https://gitlab.com/cip-project/cip-testing/testing/issues/192|#192]] to cover the docker work.
+   * Ran a BBB health check successfully with an email to the list.
+==== Thursday 2018/07/12 ====
+=== rajm ===
+   * I'm getting errors when running docker-compose on lava-docker - various packages being unavailable - I guess there's a problem with apt/sources.list at least in the image. I restarted the docker service and that appears to have resolved most of the problems, I guess it didn't like being moved from the home network to the codethink one. Still getting a failure - wget: unable to resolve host address 'images.validation.linaro.org'
+   * Ran a BBB health check successfully with an email to the list.
+==== Wednesday 2018/07/11 ====
+=== rajm ===
+   * A fresh provision for week 28 [[https://gitlab.com/cip-project/cip-testing/testing/issues/149|#149]] with a local test section. Initially had problems with the QEMU test due to / running out of space, I did a make clean in linux-cip to resolve, I'm not sure why this doesn't happen on all clean tests?
+   * Ran a BBB health check successfully with an email to the list.
+==== Friday 2018/07/06 ====
+=== rajm ===
+   * Created [[https://github.com/lucj/kernelci-docker/issues/22| kernelci-docker issue #22]] to cover the failure to download  the kernel.
+   * Noted [[https://github.com/lucj/kernelci-docker/issues/11| kernelci-docker issue #11]] I see this issue but we  don't have the problem in b@d.
+   * The build artefacts are uploaded to the kernelci-proxy container, but I think the retrieval is only happening in kernelci-frontend.
+   * Ran a BBB health check successfully with an email to the list.
+==== Thursday 2018/07/05 ====
+=== rajm ===
+   * Failing to get the kernelci docker to start with dev-start.sh- getting docker compose version is unsupported - an error which on the face of it is incorrect - as I have the latest version of docker-compose.
+   * Created version of the frontend with logging enabled in nginx and verified that the failure is happening in that container. I see that the instructions for copying an archive from a kernelci m/c to a container-ised version only copy the logs, does that mean that the main focus is on the logs?
+   * Ran a BBB health check successfully with an email to the list.
+==== Wednesday 2018/07/04 ====
+=== rajm ===
+   * I'm getting failures of b@d to start up when on wifi and also connected to work VPN I guess this is DHCP issues.
+   * I am getting 404 errors when attempting to retrieve the build artefacts from the docker image. ISTR getting similar errors in the early stages of B@D development. I will research it. This is the change to server.py in configure_singledev.sh - though even with that change I'm still getting 404's
+   * Looking at the LAVA docker.
+===== June =====
+==== Friday 2018/06/29 ====
+=== rajm ===
+   * I've successfully built a BBB image and uploaded it to the docker kernelci.
+   * Ran a BBB health check successfully with an email to the list.
+==== Thursday 2018/06/28 ====
+=== rajm ===
+   * Ran the kernelci docker image and am attempting to build with kernelci-build but am currently getting a 413 error (Request entity too large) when the build script attempts to upload the results. Reviewing the B@D settings to see where size limits get set.
+   * Tweaked my populate script so that it builds linux-cip using the latest tag.
+   * I see from the fresh provision I did yesterday that backports is now on LAVA 2018.5.post1 which probably makes merging [[https://gitlab.com/cip-project/cip-testing/board-at-desk-single-dev/merge_requests/60|MR60]] more important.
+   * Attended cip-dev IRC meeting.
+   * Ran a BBB health check successfully with an email to the list.
+   * Created a branch with a working BBB health check [[https://gitlab.com/rajm/board-at-desk-single-dev/tree/localTest|localTest]].
+==== Wednesday 2018/06/27 ====
+=== rajm ===
+   * Looking at moving B@D to containers and investigating the kernelci instances.
+   * A fresh provision for week 26 [[https://gitlab.com/cip-project/cip-testing/testing/issues/149|#149]] having missed a week because of holiday. All successfully done (with the local check).
+   * Built the latest v4.4.138-cip25 kernel for BBB on my VM and ran a local health check on it successfully with an email to the cip-testing-results list.
+   * The default BBB health check fails "Unable to open test definition 'common/dt-selftests.yaml'" I've opened [[https://gitlab.com/cip-project/cip-testing/testing/issues/191|#191]] and assume the linaro repos has changed.
+   * Returned from holiday.
+==== Monday 2018/06/25 ====
+=== SZ Lin (林上智) ===
+  * Attended TSC meeting.
+==== Thursday 2018/06/14 ====
+=== rajm ===
+   * I'll be away holidaying for a few days...
+   * However a test of the iwg20m is not sucessful. [[https://gitlab.com/cip-project/cip-testing/testing/issues/189|#189]] opened. The kernel builds ok but I'm currently unable to connect via USB to the board - there's no problems with similar cabling to the BBB.
+   * Done a fresh provision of B@D on a Windows 10 machine and verified that kernelci and the QEMU & BBB health checks run ok (using initramfs built using populate.sh)
+   * I've pushed an update to [[https://gitlab.com/rajm/board-at-desk-single-dev/tree/boxPopulate|boxPopulate]] and have verified that it runs ok (on Windows 10)
+   * Attended cip-dev IRC meeting.
+   * Ran the BBB health check successfully with the email going to the cip-testing-results list. Worked first time today.
+==== Wednesday 2018/06/13 ====
+=== rajm ===
+   * Created [[https://gitlab.com/cip-project/cip-testing/board-at-desk-single-dev/merge_requests/61|MR61]] to handle [[https://gitlab.com/cip-project/cip-testing/testing/issues/178|Issue #178]].
+   * Also ran the iwg20m heath check successfully with an email to the list.
+   * Ran the BBB health check successfully with the email going to the cip-testing-results list. A couple of failures before the working one 'send-reboot-commands timed out after 119 seconds' and 'matched a bootloader error message'.
+==== Monday 2018/06/11 ====
+=== rajm ===
+   * Updated [[https://gitlab.com/cip-project/cip-testing/board-at-desk-single-dev/merge_requests/60|MR60]] to use an Debian snapshot for LAVA 2018.4.
+   * Updated the block diagram on the features page to reflect the versions provided by the git repos and opened [[https://gitlab.com/cip-project/cip-testing/testing/issues/188|#188]] to cover consistency issues between the diagram and the rest of the text on that page.
+   * An iwg20m health check also ran successfully with the email going to the cip-testing-results list.
+   * Ran the BBB health check successfully with the email going to the cip-testing-results list.
+=== SZ Lin (林上智) ===
+  * Attended TSC meeting.
+==== Friday 2018/06/08 ====
+=== rajm ===
+   * Work on B@D Friday talk.
+   * Collaboration with cpaterson wrt the iwg20m problems and installing B@D.
+   * More updates to the wiki page.
+   * Ran the BBB health check successfully with the email going to the cip-testing-results list.
+==== Thursday 2018/06/07 ====
+=== rajm ===
+   * Updates to the wiki features page so that the version of B@D components if run from git are recorded.
+   * Created [[https://gitlab.com/cip-project/cip-testing/board-at-desk-single-dev/merge_requests/60|MR60]] to handle [[https://gitlab.com/cip-project/cip-testing/testing/issues/|#185]] I will take on board bwh's comment and revise it.
+   * Attended CIP Codethink team meeting
+   * Tried running the BBB test with a test section identical to the renesas one - gives around 10 seconds shorter run time mainly due to it not downloading a git repos, we might want to replace the BBB test or have this as an alternative - especially if testing without remote access.
+   * The remote power switcher had problems connecting to the local network which caused a lava-tool job to fail with infrastructure errors. As a consequence LAVA then set the health of the iwg20m to unknown which caused a health check to run - which also failed- because the power switch was off line! Fortunately when it came on line the HC worked and an email went to the list.
+   * Attended cip-dev IRC meeting.
+   * Ran the BBB health check successfully with the email going to the cip-testing-results list.
+=== toscalix ===
+   * CIP Codethink team meeting
+   * Linux Format published an article about CIP. Codethink's name is wrong.
+   * We have invested a significant amount of effort in making the Renesas board reliable with LAVA and we haven been able to identify the root cause of the issue. We will not dedicate further time at this point. We will wait for Renesas to make an evaluation.
+   * For OSSJ the plan is:
+      * To have a new kernel released.
+      * Talk prepared. Daniel Wagner answered the questions Agustin had about the -rt process.
+      * Docu and bugs from B@D up to date.
+      * Undetermined at this point if the kernel config evaluation of sent by Moxa can be done before the event.
+      * Codethink will participate at the CIP with Sudip and Agustin.
+   * Proposed to SZ Lin a report about the kernel stable patch review experience through cip-dev
+   * The blog post about Codethink's work will be published next Monday.
+==== Wednesday 2018/06/06 ====
+=== rajm ===
+   * Ran a successful iwg20m HC but previous ones had failed - no apparent software change between the non-working and working tests.
+   * A fresh provision for week 23 [[https://gitlab.com/cip-project/cip-testing/testing/issues/149|#149]] some issues with connectivity (git clone timing out) and lack of disk space causing QEMU to fail but eventually both worked.
+   * Tried running the iwg20m health check with the Renesas core-image-minimal-iwg20m.cpio.gz but that exhibits the same timeout on starting kernel.
+   * Ran the BBB health check successfully with the email going to the cip-testing-results list.
+==== Friday 2018/06/01 ====
+=== rajm ===
+   * Comparing a test of iwg20m with the same kernel one working one not working [[https://paste.baserock.org/gicaneyezu|shows]] that the test doesn't get as far as 'Uncompressing Linux' so the test is failing early in initramfs. In that batch of tests 3 worked and one failed.
+   * It appears that specifying the initramfs fails to work with U-Boot so a u-boot header needs to be added.
+   * Built iwg20m kernel with EXT4 support and am successfully booting via u-Boot and am trying to provoke it to hang on boot as it does in LAVA. So far in this environment boots are 'just working' - not the desired outcome!
+   * Ran the BBB health check successfully with the email going to the cip-testing-results list.
+   * 1:1 with toscalix.
 ===== May =====
+==== Thursday 2018/05/31 ====
+=== rajm ===
+   * Had issues with the remote power switch getting a new address more frequently with DHCP.
+   * Attended team meeting.
+   * Discussed with bwh my problems with the iwg20m - it needs to be built with EXT4 support or tested using our initramfs.
+   * Attended cip-dev IRC meeting.
+   * Ran the BBB health check successfully with the email going to the cip-testing-results list.
+=== SZ Lin (林上智) ===
+  * Attended cip-dev IRC meeting.
+==== Tuesday 2018/05/29 ====
+=== rajm ===
+   * Reply from Neil Williams about the QEMU locking up issue. I've replied to it with what information I can provide but as the device has now health checked (and the VM rebooted) it's hard to provide the full information.
+   * Another failure from the iwg20m - in order to get an attempt working I had to re-power cycle the remote power switcher so that it reset its IP address (addresses having changed in the new office setup).
+   * Ran the BBB health check successfully with the email going to the cip-testing-results list.
+   * Successfully setup equipment in new office.
+=== toscalix ===
+   * Sent a couple of interesting links to the cip-testing-results ML for the new stable reviewers.
+==== Monday 2018/05/28 ====
+=== toscalix ===
+   * Created a report for the CIP TSC meeting. I cannot attend today.
+   * Worked on issue [[https://gitlab.com/cip-project/cip-testing/testing/issues/187|#187]].
+   * Worked on the OSSJ presentation.
+=== SZ Lin (林上智) ===
+  * Attended TSC meeting.
+==== Friday 2018/05/25 ====
+=== rajm ===
+   * Packing away equipment for the office move this weekend.
+   * Tested script to populate a provisioned instance with HCs and build initramfs and kernel (built for BBB) [[https://gitlab.com/cip-project/cip-testing/testing/issues/150|#150]].
+   * Ran the BBB healthcheck successfully with the email going to the cip-testing-results list.
+==== Thursday 2018/05/24 ====
+=== rajm ===
+   * No response yet from lava-users about the QEMU problem - however the QEMU health check ran ok today, but I'll see if the bug needs reporting once there is feedback.
+   * Attended cip-dev IRC meeting.
+   * Anh The reported that the kernelci build script doesn't copy uImage onto the webserver. I checked out build.py and edited patterns in that file which then copied over uImage however the kernelci webserver doesn't report it.
+   * Ran the BBB healthcheck successfully with the email going to the cip-testing-results list.
+=== SZ Lin (林上智) ===
+  * Attended cip-dev IRC meeting.
+==== Wednesday 2018/05/23 ====
+=== rajm ===
+   * Sent email to the lava-users list asking about the 'maybe a bug in LAVA' problem mentioned below.
+   * Created a skeleton script for the configuration of a new instance once the VM is provisioned.
+   * 1:1 with toscalix.
+   * Done a fresh provision for week 21 of [[https://gitlab.com/cip-project/cip-testing/testing/issues/149|#149]] all 3 health checks working first time.
+   * Runs the BBB healthcheck successfully with the email going to the cip-testing-results list.
+   * Replied to Anh The's support request email.
+   * Tried U-Boot of the iwg20m using /dev/mmcblk[0-2]p2 as root device. k0 gives a kernel panic the others hang waiting for the root device. Get similar results  with the yocto build kernel.
+   * The QEMU health check is failing to start - or maybe the error at the head of the page "Unable to parse invalid logs: This is maybe a bug in LAVA that should be reported." is preventing the display of the output? Asked in IRC #linaro-lava channel on Friday but I didn't see any response. Still getting this error after rebooting the VM (and also the host laptop).
+==== Tuesday 2018/05/22 ====
+=== toscalix ===
+   * Sent to Mae the content of the post about Codethink activity. I forgot to send it last week.
+   * I promoted the recently published blog post through twitter and linkedin.
 ==== Friday 2018/05/18 ====
@@ Line 9: / Line 424: @@
 === rajm ===
+   * I also closed [[https://gitlab.com/cip-project/cip-testing/testing/issues/183|#183]] as it belonged with #184.
    * 1:1 with toscalix.
    * BBB health check runs and email sent to list (IWG20M HC fails)
@@ Line 31: / Line 447: @@
    * cip-dev meeting. The best  one so far.
+=== SZ Lin (林上智) ===
+  * Attended cip-dev IRC meeting.
 ==== Wednesday 2018/05/16 ====

Wiki

User Tools

Site Tools

Differences

Page Tools