CI builds failing due to lack of memory space

Tags: #<Tag:0x00007f2994cfea30> #<Tag:0x00007f2994cfe8f0> #<Tag:0x00007f2994cfe7b0>

Hi all,

I am releasing the OpenMRS Platform in the alpha version. For the previous 3 days, I have had successful builds. It is today that I am getting problems for the platform builds. I get such error logs as related to memory insufficiency. Is there a way this can be handled so we have the Platform released out? Thanks.

Regards

cc: @cintiadr, @mogoodrich, @dkayiwa, @c.antwi

1 Like

cc @mozzy did you get such error some time back?

Hi.

If you’d like me to take a look on a CI build, I need the link to it.

1 Like

This was resolved @cintiadr, Thanks. I now have successful builds. Not clear though what could have caused this log https://ci.openmrs.org/browse/OP-OPM7-3/log.

So I can see that this job ran in yokobue agent at 29 Aug 2019, 5:35:42 PM (UTC time).

The messages are pretty helpful, error: file write error (No space left on device) and fatal: write error: No space left on device, which clearly indicate that yokobue ran out of disk (no idea what lead you to think that we had memory problems).

So, getting that data for the last 7 days:

28%20pm

Indeed, somehow 9:00am on the 29th (UTC) something started using a lot more disk than usual (and it went back to normal around 10pm same day).

Now, I did receive an alarm, but those are the ones I’m just going to check in a couple of days, if they are still complaining. So I didn’t really do anything, and it self resolved before I checked it.

Again, this investigation is only possible because you linked the CI build. It’s impossible to investigate without a proper link and all the logs.

1 Like