If it helps I would like to be part of this. I have experience managing Jira and confluence. I’m not a expert but, I can try to help as much as possible like the others.
CPU spike or memory spike?
CPU. Memory spikes are almost always attributed to ID dashboard.
Now the heap has just 11% free memory, in any moment the garbage collector is going to kick-in and CPU will spike. I definitely think it’s a memory problem (maybe a memory leak, maybe not).
You’re more than welcome to have at it if you like
Thank you for your offer, but if you give tips, great – but I’d rather not give people who are so new to the community access to a production server and/or administrative access. If you’re established it’s one thing, but you’re too new @chagara…I really appreciate the offer and would welcome your tips on tuning our performance – I’m very careful usually even @lluismf having access would be temporary to fix the issue(s) we’re having…
I can’t stop JIRA and modify memory parameters. Also, the server has just 8Gb so giving JIRA an extra Gb could affect other applications running on it. 64 bit Linux can address Terabytes of RAM, RAM is very cheap these days => buy more RAM and give it plenty to JIRA, Confluence, etc.
We’re on a VPS donated to us by IU running on XSEDE. When I said we have what we have, I was serious.
Maybe Atlassian Cloud would like to host us and help with the migration
One of the plugins we use for the wiki is not available on Atlassian Cloud. The one which pulls the latest posts here on Talk. I’m not sure entirely. I know the Bahmni team wanted that widget.
JIRA seems to have performance problems running on virtualized systems (I can’t locate the document now, but I read it). So, our hands are tied. Can’t add more RAM, and the server is not ours.
Perhaps @chagara and/or @lluismf could do a deep dive on investigating: (a) what exactly would stop us from hosting on Atlassian Cloud (b) what are all the plugins we have, and which are truly needed © what size AWS/DigitalOcean instance would it take to make this run well (i.e. what is the actual cost)
I can do what the research Darius suggested. The atlassian Cloud is decent but, they are limitations. As far as I know DigitalOcean offers a discount or free VM for open source projects so that can be a route. If you guys want me to do the research I can do so.
I can list the plugins/addons and then ask the community to see which ones can be disabled. Hopefully this will make startup faster.
We already have infrastructure on Digital Ocean. Digital Ocean is also virtualized, and things get pricey as the amount of RAM rises. OpenMRS Talk runs on DigitalOcean, as does our Telegram Bot.
Somebody who isn’t me me can do the pricing and research Both AWS and DigitalOcean are upfront about their pricing.
what options are possible then?
I don’t know right now. We should investigate what plugins we’re using and prune them.
@chagara, you should explore what it would require for us to host on Atlassian Cloud and/or AWS/DigitalOcean.
I would expect the answer to include:
- to host on Atlassian we’d need to remove plugins xyz, and pay $xyz per month (or get them to donate it)
- to host on AWS we’d need a server of size xyz, which will cost $xyz per month
Right now we have 1GB of RAM dedicated to JIRA (if I’m reading the image above correctly).
To double that on Digital Ocean would cost $20/mo.
EDIT: Apparently non-profit pricing is available.