Closed skandyla closed 6 years ago
@skandyla, thanks for the report.
This error also happens when the salt-ssh output is too large. For example replacing a 254K file using file.managed
(resulting in a huge diff being printed) triggered it.
Salt: 2015.5.3
Python: 2.7.6 (default, Jun 22 2015, 17:58:13)
Jinja2: 2.7.2
M2Crypto: 0.21.1
msgpack-python: 0.3.0
msgpack-pure: Not Installed
pycrypto: 2.6.1
libnacl: Not Installed
PyYAML: 3.10
ioflo: Not Installed
PyZMQ: Not Installed
RAET: Not Installed
ZMQ: Not Installed
Mako: Not Installed
Tornado: Not Installed
Debian source package: 2015.5.3+ds-1trusty1
Hi, this error also very hard to debug. Since salt-ssh doesn't report which module caused it.
For example, what I start getting recently:
# time salt-ssh remotesrv.com state.highstate
[ERROR ] JSON Render failed for:
Permission denied, no authentication information
[ERROR ] No JSON object could be decoded
remotesrv.com:
real 7m31.709s
user 0m15.327s
sys 0m3.276s
# time salt-ssh remotesrv.com state.highstate
remotesrv.com:
Summary for remotesrv.com
--------------
Succeeded: 164
Failed: 0
--------------
Total states run: 164
real 0m58.189s
user 0m12.150s
sys 0m1.597s
Debug logging doesn't help.
164 states looks a lot like it could be the same problem as mine (output too large).
Especially since the second highstate run is very likely to contain less output than the first.
Actually the size of the output doesn't matter. It fails with less count of states too:
mysrv # time salt-ssh -i --wipe remotesrv.com state.highstate
[ERROR ] JSON Render failed for:
Permission denied, no authentication information
[ERROR ] No JSON object could be decoded
remotesrv.com:
real 2m29.970s
user 0m10.925s
sys 0m1.720s
mysrv # time salt-ssh -i --wipe remotesrv.com state.highstate
remotesrv.com:
Summary for remotesrv.com
-------------
Succeeded: 81
Failed: 0
-------------
Total states run: 81
real 0m36.515s
user 0m8.605s
sys 0m1.336s
It seems, that two my latests comments it's another issue, caused by deploying to ec2 nodes with sudo.
remotesrv.com:
host: 111
user: centos
sudo: true
I've tried to deploy to DigitalOcean with the same settings, but without sudo - and it's working good.
I don't know is it actual, but I got same issue. In my case I tried to rewrite configuration file with 650 lines(mostly comments) by 6 lines of pure configuration, it looks like it can't return diff. Of course this is not a normal behavior but I'm resolved this issue via file cleaning
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.
If this issue is closed prematurely, please leave a comment and we will gladly reopen the issue.
Hi, this issue is based on the following discussion https://groups.google.com/forum/?pli=1#!topic/salt-users/elKOx-T0yFM
In some cases, when we install pkg and create the config for it, salt reporting with
But actually state executes normal.
Actual Behavior
Just to show, that screen package is absent at my system:
The first run:
Second:
Steps to Reproduce Issue
salt-ssh example.com -r 'yum -y remove screen; rm /etc/screenrc'
salt-ssh example.com state.sls tools.screen
Versions Report
\ Provided by running
salt --versions-report
**Relevant Logs, Configs, or States
cat tools/screen/init.sls
The contents of the screenrc doesn't matter. I also tried to provide config with contents_pillar:
- contents_pillar: screen:screenrc
without any success.roster is trivial:
The issue reproduces on every of my centos7 srvs, and doesn't reproduces on ubuntu\debian srvs. Also, this is not screen-related issue. I've got the same issue at my
chrony
module.