Compatibility layer for EESSI version 2026.06#235
Conversation
|
bot: build repo:eessi.io-2025.06-compat arch:x86_64/generic |
|
bot: build repo:eessi.io-2025.06-compat instance:eessi-bot-mc-aws for:arch=x86_64/generic |
|
New job on instance
|
|
bot: build repo:eessi.io-2025.06-compat instance:eessi-bot-mc-aws for:arch=x86_64/generic |
|
New job on instance
|
|
bot: build repo:eessi.io-2025.06-compat instance:eessi-bot-mc-aws for:arch=x86_64/generic |
|
New job on instance
|
|
bot: build repo:eessi.io-2025.06-compat instance:eessi-bot-mc-aws for:arch=x86_64/generic |
|
New job on instance
|
|
bot: build repo:eessi.io-2025.06-compat instance:eessi-bot-mc-aws for:arch=x86_64/generic |
|
New job on instance
|
|
We decided today to ingest the compat layer as #231 is necessary in the near future but not a strict blocker for |
|
bot:status last_build |
|
This is the status of all the
|
|
I really don't understand what I'm seeing in the staging PR... I mean, I see only one uploaded tarball in this PR. #235 (comment) and #235 (comment) don't seem to have been uploaded for some reason. Yet in the staging PR I see three tarballs, with names ranging from What I also don't understand is that our build command is: I.e. the target repo is |
|
It seems the AWS cluster uploaded that strange And at the end of it, it seems to have crashed (when uploading the metadata file?): This may be why it didn't proceed to the others (?) |
|
Ok, let me first just try to redeploy... |
|
Crash seems to be consistent. I'm thinking: it may be related to the fact that the bot is trying to add to an existing PR comment that got deleted - and that might mess up the entire state. I mean: I mean upload |
|
Renamed the two items it would normally upload, let's see if this prevents it from uploading these... |
|
Even worse: I guess it doesn't actually check the directory, just reconstruct the name based on the metadata and just assume the file exists. Ok, next option: see if I can make the bot believe this was a failed build so that it will skip it... |
|
Oof, there's another one as well: leads to the same crash. Will try the same workaround. |
|
Btw, the workaround I do is: and then modify the |
|
Ugh, there are more... It sees 17 successful jobs a number of which seem to be |
|
Ok, did the same for jobs 160219 to 160228... let's see how it goes now. |
|
Now we're getting somewhere. #235 (comment) just uploaded, the other one is on the way. I'll have to go for dinner. But I fully expect these to show up in the staging PR on the next pass by the Stratum 0. |
|
@casparvl Since staging PR is merged, we should merge this as is? |
Still WIP, but testing some new functionality already...