Diagnostic snapshot doesn't include content from slave-nodes

[Alexander Rubtsov]

Release: Mirantis OpenStack 4.1

Expected result:
diagnostic snapshot include configuration files (/etc/...) and output of some command from slave-nodes

Actual result:
- directories in ./node-x.domain.tld/etc/ are empty
- txt files in ./node-x.domain.tld/commands/ with the following content:
===== COMMAND =====: <command>
===== RETURN CODE =====: None
===== STDOUT =====:
None
===== STDERR =====:
None

[Alexander Rubtsov (arubtsov)]

Diagnostic snapshot is attached

[Dmitry Pyzhov (dpyzhov)]

I can't reproduce the issue, it seems to be local environment issue

[Mike Scherbakov (mihgen)]

I'm setting Incomplete for now. I hope we can get more info about an env to confirm this bug. Right now, it's likely that there were no particular files on slaves or commands, so it returned nothing, or some other issue not directly related to shotgun library, which does logs collection at the moment.
We need more troubleshooting actions to check manually what really happen when we issue those commands on slaves.

Please try to reproduce the issue, and check commands by manually running them on env. Also, please ping us at #fuel-dev, so we can collaborate in run time.

[Mike Scherbakov (mihgen)]

There were no activity on this bug for a while. If you reproduce this issue, please reopen it. Closing as Invalid for now.

[Bogdan Dobrelya (bogdando)]

In logs attached to https://bugs.launchpad.net/fuel/+bug/1373569 there are *still* empty commands.txt files in snapshot, looks like this bug wasn't fixed completely

[Tatyanka (tatyana-leontovich)]

{u'build_id': u'2014-10-03_00-00-10', u'ostf_sha': u'f238ecaef0cad7d31aca7e6af2f2523d8fcdc9e7', u'build_number': u'542', u'auth_required': True, u'nailgun_sha': u'4460732478cfcabc6f656d9d2e41cab4de64410d', u'production': u'docker', u'api': u'1.0', u'fuelmain_sha': u'0a7cfb07241741dba3ecbd14210512b9eb1d88ae', u'astute_sha': u'65a5a0781d314bdcc090a61a599a8abf95989fc1', u'feature_groups': [u'mirantis'], u'release': u'6.0', u'release_versions': {u'2014.2-6.0': {u'VERSION': {u'build_id': u'2014-10-03_00-00-10', u'ostf_sha': u'f238ecaef0cad7d31aca7e6af2f2523d8fcdc9e7', u'build_number': u'542', u'api': u'1.0', u'nailgun_sha': u'4460732478cfcabc6f656d9d2e41cab4de64410d', u'production': u'docker', u'fuelmain_sha': u'0a7cfb07241741dba3ecbd14210512b9eb1d88ae', u'astute_sha': u'65a5a0781d314bdcc090a61a599a8abf95989fc1', u'feature_groups': [u'mirantis'], u'release': u'6.0', u'fuellib_sha': u'df3377534acd70389b551025539152039e260f5c'}}}, u'fuellib_sha': u'df3377534acd70389b551025539152039e260f5c'}

[Tatyanka (tatyana-leontovich)]

works fine on {"build_id": "2014-10-03_00-00-10", "ostf_sha": "f238ecaef0cad7d31aca7e6af2f2523d8fcdc9e7", "build_number": "542", "auth_required": true, "api": "1.0", "nailgun_sha": "4460732478cfcabc6f656d9d2e41cab4de64410d", "production": "docker", "fuelmain_sha": "0a7cfb07241741dba3ecbd14210512b9eb1d88ae", "astute_sha": "65a5a0781d314bdcc090a61a599a8abf95989fc1", "feature_groups": ["mirantis"], "release": "6.0", "release_versions": {"2014.2-6.0": {"VERSION": {"build_id": "2014-10-03_00-00-10", "ostf_sha": "f238ecaef0cad7d31aca7e6af2f2523d8fcdc9e7", "build_number": "542", "api": "1.0", "nailgun_sha": "4460732478cfcabc6f656d9d2e41cab4de64410d", "production": "docker", "fuelmain_sha": "0a7cfb07241741dba3ecbd14210512b9eb1d88ae", "astute_sha": "65a5a0781d314bdcc090a61a599a8abf95989fc1", "feature_groups": ["mirantis"], "release": "6.0", "fuellib_sha": "df3377534acd70389b551025539152039e260f5c"}}}, "fuellib_sha": "df3377534acd70389b551025539152039e260f5c"}

[Mike Scherbakov (mihgen)]

As it was verified by QA (see comment #8), should not we close it as Invalid?

[Tatyanka (tatyana-leontovich)]

It was verified a long time ago, and I can not see such problems in snapshots, created during nightly tests runs - so move to invalid now

[Bogdan Dobrelya (bogdando)]

Please elalborate long time ago

[Bogdan Dobrelya (bogdando)]

According to https://bugs.launchpad.net/fuel/+bug/1303754/comments/5, it was reproduced a month ago for a last time

[Dima Shulyak (dshulyak)]

Something like this:

None[root@nailgun fuel-snapshot-2014-11-03_10-01-44]# cat node-2.test.domain.local/commands/blkid_o_list.txt
===== COMMAND =====: blkid -o list
===== RETURN CODE =====: None
===== STDOUT =====:
None
===== STDERR =====:

Can be present in snapshot in case there is connectivity problem with vm, easy to reproduce by suspending vm, for example.