-
Notifications
You must be signed in to change notification settings - Fork 49
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Some suspected failings when a CN falls over #84
Comments
Hi @magnayn, which sdcadm version are you using?. Hostvolumes are not needed anymore and should be removed by |
[root@headnode (Osney) ~]# sdcadm --version doing a selfupdate to 1.11.2 fixes it. D'oh! On Fri, Jun 17, 2016 at 5:01 PM, Pedro Palazón Candel <
|
hm - that said, after upgrading and sdcadm post-setup docker : [root@headnode (Osney) ~]# sdcadm update docker /opt/smartdc/sdcadm/lib/sdcadm.js:720 |
I think that's b/c you removed the server from AdminUI w/o deleting the instances from SAPI. Mind to tell me what's the output of:
|
Also, same thing for
|
Ah - ok - I wasn't aware I needed to do non-adminui stuff
|
|
Ok, run: Then, |
Hmm - there is an error recorded in the update [root@headnode (Osney) ~]# sdcadm experimental update-other sdcadm experimental: error: socket hang up [root@headnode (Osney) ~]# sdc-cnapi { On Fri, Jun 17, 2016 at 5:18 PM, Pedro Palazón Candel <
|
Can you get the output of |
hmm [root@headnode (Osney) ~]# sdcadm health /opt/smartdc/sdcadm/lib/sdcadm.js:720 On Fri, Jun 17, 2016 at 5:22 PM, Pedro Palazón Candel <
|
Try |
from this list, hostvolume-SFO server was destroyed in adminui [root@headnode (Osney) ~]# sdc-role list On Fri, Jun 17, 2016 at 5:25 PM, Pedro Palazón Candel <
|
Mind to give a try to: |
health-check and 'sdcadm update docker' are now happy [root@headnode (Osney) ~]# sdcadm update docker experimental update-other isn't, but I don't know if that is significant [root@headnode (Osney) ~]# sdcadm experimental update-other FWIW
On Fri, Jun 17, 2016 at 5:29 PM, Pedro Palazón Candel <
|
It looks like something isn't Ok either for VMAPI or SAPI there. Could you take a look at those service logs - within the vms - and see if there's any error? |
I had to leave for the weekend; coming back I found svc:/manta/application/binder:default (Joyent DNS-ZooKeeper Service) (log was empty) so I've restarted it On Fri, Jun 17, 2016 at 5:37 PM, Pedro Palazón Candel <
|
I had a CN fall over, and this seemed to cause docker provisioning to fail:
When trying to update things just to check, it also failed :
I destroyed the server in adminui. Docker provisioning now works - hwever updating still doesn't
Feels like there is some behaviour that assumes servers are always alive/up.
The text was updated successfully, but these errors were encountered: