action #46967
closed[functional][u][tools] warning in bootloader_zkvm: Calling Net::SSH2::Channel::readline in non-blocking mode is usually a programming error
Description
Observation¶
openQA test in scenario OpenQA::Schema::Result::TestSuites=HASH(0xaeb1b28) fails in
bootloader_zkvm
[2019-02-01T10:03:04.876 CET] [debug] Connection to root@s390p8.suse.de established
Calling Net::SSH2::Channel::readline in non-blocking mode is usually a programming error at /usr/lib/os-autoinst/backend/baseclass.pm line 1221.
First time seeing this in Beta2 Candidate of SLE15-SP1, restarting worked though
Workaround¶
- restarting the test
Reproducible¶
Fails since (at least) Build 158.4 (current job)
Expected result¶
Last good: 157.1 (or more recent)
Further details¶
Always latest result in this scenario: latest
UPDATE¶
bootloader_zkvm failures are not related to this warning. This warning comes from perl-Net-SSH2 version 0.66 (commit ad5f9f0 ("Warn about readline being called in non-blocking mode"). I guess 157.1 (last good) also had this warning, just didn't fail due some other reason. I'll fix this issue during fixing #45863.
Updated by mgriessmeier almost 6 years ago
another occurence: https://openqa.suse.de/tests/2429791/file/autoinst-log.txt
Updated by mgriessmeier almost 6 years ago
actually this also occurs in tests which are successful, so it's more a warning
Updated by pvorel almost 6 years ago
It's possible to debug it if you set debug mode for SSH2: Net::SSH2->new(debug => 1);
Beware it takes a lot data.
Updated by okurz almost 6 years ago
- Subject changed from [functional][tools] test fails in bootloader_zkvm: Calling Net::SSH2::Channel::readline in non-blocking mode is usually a programming error to [functional][u][tools] test fails in bootloader_zkvm: Calling Net::SSH2::Channel::readline in non-blocking mode is usually a programming error
- Target version set to future
@mgriessmeier please always use a team-assignment like [u] or [y]. As you know, we want the teams to be more independant so we should rely less on "catch-all" approaches like people like me to crosscheck the tickets that fall between the silos ;)
Updated by szarate almost 6 years ago
- Related to action #46919: [functional][u][svirt][sporadic] auto_review:"IO::Socket::INET: connect: Connection timed out" added
Updated by pvorel over 5 years ago
BTW I use Net::SSH2::Channel for another serial connection (over nc/virsh console call) for svirt serial terminal in blocking mode in this PR.
https://github.com/os-autoinst/os-autoinst/pull/1109 (#45863)
I hope I'll find a time to address this problematic connection in it as well.
Updated by jlausuch over 5 years ago
@mgriessmeier: can you please change the version of this ticket to "Planned Kernel&Network"?
Updated by okurz over 5 years ago
@jlausuch I do not know who added you in that weird way to the projects however I added you now to https://progress.opensuse.org/projects/qa/settings/members as manager so you should be able to do everything yourself now :)
Updated by pvorel over 5 years ago
- Assignee set to pvorel
- Target version changed from 454 to 445
- Parent task set to #45863
Actually, this must be solved during fixing svirt serial implementation in #45863.
Updated by pvorel over 5 years ago
- Subject changed from [functional][u][tools] test fails in bootloader_zkvm: Calling Net::SSH2::Channel::readline in non-blocking mode is usually a programming error to [functional][u][tools] warning in bootloader_zkvm: Calling Net::SSH2::Channel::readline in non-blocking mode is usually a programming error
Updated by jlausuch over 5 years ago
okurz wrote:
@jlausuch I do not know who added you in that weird way to the projects however I added you now to https://progress.opensuse.org/projects/qa/settings/members as manager so you should be able to do everything yourself now :)
Thanks :)
Updated by pvorel over 5 years ago
- Related to action #45863: [s390x][svirt][ltp] Fix serial terminal console implementation for svirt backend and use it's output added
Updated by pvorel over 5 years ago
- Related to action #49394: [s390x][svirt] Print content of kernel log and serial console into "Serial output" field added
Updated by okurz over 5 years ago
- Category changed from 132 to Feature requests
Updated by SLindoMansilla about 5 years ago
- Priority changed from Normal to High
Updated by mkittler almost 5 years ago
- Category changed from Feature requests to Regressions/Crashes
- Status changed from New to Resolved
- Assignee set to mkittler
- Target version changed from future to Done
This should be fixed with https://github.com/os-autoinst/os-autoinst/pull/1239 and I haven't seen this log message anymore. Recent tests in the same scenario are not failing anymore, too.