[Gluster-infra] git.gluster.org sick?

Vijay Bellur vbellur at redhat.com
Sun Dec 20 06:49:04 UTC 2015


On 12/20/2015 12:04 AM, Vijay Bellur wrote:
> On 12/19/2015 04:22 PM, Dan Lambright wrote:
>>
>>
>> ----- Original Message -----
>>> From: "Kaleb Keithley" <kkeithle at redhat.com>
>>> To: "gluster-infra" <gluster-infra at gluster.org>
>>> Sent: Saturday, December 19, 2015 1:14:38 PM
>>> Subject: [Gluster-infra] git.gluster.org sick?
>>>
>>>
>>> I'm getting "fatal: internal server error"when I try to clone from
>>> git.gluster.org
>>
>> Also affects patch submittal, help needed here..
>>
>> [root at rhs-cli-11 glusterfs]# ./rfc.sh
>>
>
> Looking into this.
>

Observed two problems:

1. glusterfs.git requiring a git gc.

2. gerrit service hogging close to 50% of avaialble memory.

Completed git gc & restarted gerrit. It seems to be working fine now - 
git pull & clone operations completed successfully.

We probably would need to:

1. run git gc through cron - maybe once every 2 weeks or so.

2. Add more memory to the VM that hosts r.g.o if we cannot determine 
soon why gerrit seems to progressively increase memory utilization over 
time.

Regards,
Vijay



More information about the Gluster-infra mailing list