fuel snapshot hangs at "Generating dump..."
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Fuel for OpenStack |
Fix Committed
|
High
|
Georgy Kibardin | ||
8.0.x |
In Progress
|
High
|
MOS Maintenance | ||
Mitaka |
Fix Released
|
High
|
Georgy Kibardin |
Bug Description
Detailed bug description:
when run 'fuel snapshot' just "Generating dump..." appears.
at shotgun log can see:
2016-07-19 08:48:22 DEBUG 9018 (driver) Getting local file: cp -r /var/log/
2016-07-19 08:48:22 DEBUG 9018 (utils) Trying to execute command: mkdir -p "/var/dump/
2016-07-19 08:48:22 DEBUG 9018 (utils) Trying to execute command: cp -r "/var/log/
2016-07-19 08:48:22 DEBUG 9018 (utils) Trying to execute command: tar cJvf /var/dump/
2016-07-19 08:50:00 DEBUG 9018 (utils) Trying to execute command: rm -r /var/dump/
2016-07-19 08:50:00 INFO 9018 (cli) Snapshot path: /var/dump/
but directory even does not exist:
[root@nailgun ~]# ll /var/dump
ls: cannot access /var/dump: No such file or directory
Steps to reproduce:
- install cluster by dos.py
- run 'fuel snapshot'
Expected results:
- execution finished with no errors
- snapshot created in /var/dump/
Actual result:
- execution hangs
- if to download config by 'fuel snapshot --conf > dump_conf.yaml'
and adjust timeout field to 120 and run 'fuel snapshot' it fails with:
[root@nailgun ~]# fuel snapshot < dump_conf.yaml
Generating dump...
Traceback (most recent call last):
File "/usr/bin/fuel", line 10, in <module>
sys.
File "/usr/lib/
return func(*args, **kwargs)
File "/usr/lib/
parser.parse()
File "/usr/lib/
actions[
File "/usr/lib/
method(params)
File "/usr/lib/
directory=
File "/usr/lib/
download_handle = urllib.
File "/usr/lib64/
return _opener.open(url, data, timeout)
File "/usr/lib64/
response = self._open(req, data)
File "/usr/lib64/
'_open', req)
File "/usr/lib64/
result = func(*args)
File "/usr/lib64/
return self.do_
File "/usr/lib64/
h = http_class(host, timeout=
File "/usr/lib64/
self.
File "/usr/lib64/
raise InvalidURL(
httplib.InvalidURL: nonnumeric port: '8000Dump is timed out'
Reproducibility:
several times
Workaround:
n/a
Impact:
can't create diagnostic snapshot
Description of the environment:
VERSION:
feature_groups:
- mirantis
production: "docker"
release: "8.0"
api: "1.0"
build_number: "570"
build_id: "570"
fuel-nailgun_sha: "558ca91a854cf2
python-
fuel-agent_sha: "658be72c4b42d3
fuel-
astute_sha: "b81577a5b7857c
fuel-library_sha: "c2a335b5b725f1
fuel-ostf_sha: "3bc76a63a9e7d1
fuel-mirror_sha: "fb45b80d7bee58
fuelmenu_sha: "78ffc73065a967
shotgun_sha: "63645dea384a37
network-
fuel-upgrade_sha: "616a7490ec7199
fuelmain_sha: "d605bcbabf3153
Additional information:
shotgun log attached
also if to adjust timeout in config and dump fails to create at Fuel Ui I can see:
19-07-2016
08:59:30Dump is timed out
08:49:23Dump is timed out
08:47:08Dump is timed out
08:12:36Dump is timed out
tags: | added: area-python |
no longer affects: | fuel/newton |
Changed in fuel: | |
assignee: | Fuel Sustaining (fuel-sustaining-team) → Georgy Kibardin (gkibardin) |
Changed in fuel: | |
status: | Incomplete → In Progress |
Shotgun logs says that snapshot collection has been completed successfully. But it disappeared afterwards. Would be greate to take a look at other logs from the master node, such as:
/var/log/nailgun
/var/log/astute
/var/log/syslog (just in case)