Any ideas to get owncloud to reboot/restart regularly?
I am using the UCS vm version 10.2.1 and it runs fine for about a week or two, and then it appears to be very slow and sluggish and eventually stops responding.
Typically we check the vm and see how the cpu usage is, and it was normal, just complete non-responsive.
A restart of the vm fixes it.
It would be acceptable to Schedule a weekly restart during off-peak times (midnight to 5am), I am just wondering if that is a owncloud setting, or a UCS setting, or a VSphere ESXi schedule.
Perhaps the 10.3 may fix this, but I do not have the skillset to run a manual update.
If you really want to reboot the VM, you’ll have to do it through UCS or your hypervisor (ESXi).
What appears slow and sluggish? The WebUI? ownCloud WebUI or UCS appliance WebUI? Can you still login via ssh?
Where exactly and how do you do it? Because on a UCS appliance there could be problems with the ownCloud docker container, other docker containers that are part of the stack, the database, the docker service, any other part of the system.
Typically it can be pretty hard to debug performance problems if docker is involved as there are so many different stacks.
Have you considered memory issues (not enough for example), storage throughput issues and network throughput issues? Monitor these metrics and have a look if there coincides a spike during this time and I’m pretty sure you’ll find the culprit.
Performance issues can be very hard to debug, I wish you the best of luck!
I installed the “Univention-App-owncloud-ESX.ova” UCS vm straight out-of-the-box as downloaded and documented by owncloud (I have no idea what docker is).
There is plenty of horsepower across the board on the VSphere physical host, enough such that cpu/mem/storage is never an issue for any vm, regardless of what it is used for.
I really don’t have the skillset to dig into a sluggish owncloud, that’s why I used the UCS vm. A weekly restart is sufficient, so I don’t have to do it manually.
this looks to me like an undersized VM as well. If applications like ownCloud are getting “sluggish” and a restart is solving it i think the operating system is starting to “swapping”.
The owncloud vm is not being used in production, so the number of users is exactly 3 people that know about it.
Then anywhere from 10-14 days, all web experience with the owncloud becomes sluggish and within hours it just goes not-responding.
The 1 vCPU and the 1.5G vRAM are pre-existing vm params chosen by the uCS template. Are you recommending that I change those?
We could go into the details of the hardware of the physical host, believe me when I say we have plenty of horsepower, a HPE ProLiant with loads and loads of cores, cache, and DDR3 ram galore…
personally the first thing it would do is indeed to increase the memory. 1.5 GB looks to me quite low when reading that the UCS appliance has a quite complex setup:
As an alternative maybe https://help.univention.com/c/ucs could be another support possibility as the appliance is based on UCS where deeper knowledge about the UCS is required for such kind of issues and where i’m not sure other users in this forums have them.
