Hello Edward,

I found an old issue (#16) <https://github.com/gocd/gocd/issues/16> which
talks about this. It seems to suggest that jetty itself is writing it to a
temporary file in that directory. I'm not entirely sure why it is not using
java.io.tmpdir. So, I don't think changing that will help, but you can try.
The instructions to change it are here
<https://docs.gocd.org/current/advanced_usage/other_config_options.html#system-properties>.
You'd need to add something like:

GO_SERVER_SYSTEM_PROPERTIES="-Djava.io.tmpdir=/var/data/tmp"

in /etc/default/go-server

Cheers,
Aravind


On Tue, Feb 6, 2018 at 9:03 AM, Edward Seager <[email protected]>
wrote:

> Hi Aravind,
>
> 1. Yes it does. I also discovered the usage was quite high due to the
> flyweight directory being quite full from git checkouts. I didn't plan for
> this because the docs does not mention this behaviour, so it might be worth
> mentioning that the material sizes should be considered when planning
> server installations.
>
> 2.  "java.io.tmpdir": "/tmp",
>
> How would I go about reconfiguring that? I managed to catch the upload in
> the act and it looked like the directory being used was actually
> /var/lib/go-server/work/jetty-0.0.0.0-8153-cruise.war-_go-any
>
> I've managed to resolve the issue now by increasing the size of the
> partition.
>
>
>
> On Wednesday, January 31, 2018 at 12:38:28 PM UTC, Aravind SV wrote:
>>
>> Hello Edward,
>>
>> 1. Does the spike/warning go away after the artifact is uploaded?
>>
>> 2. When you go to http(s)://your-server*/go/api/support* and look under
>> "Runtime Information" and then "System Properties", what is the value you
>> see for "java.io.tmpdir"?
>>
>> I'm not sure where the file is being (temporarily) uploaded to. I just
>> wondered whether it is the tmpdir, and if so, you might be able to
>> configure it to "/var/data/tmp" or something and see if it helps.
>>
>> Cheers,
>> Aravind
>>
>>
>>
>>
>> On Wed, Jan 31, 2018 at 5:43 AM, Edward Seager <[email protected]>
>> wrote:
>>
>>> I've figured out what is causing this... a job is uploading a 3GB
>>> artifact which is causing the spike in the disk usage.
>>> The artifacts directory is on a separate partition /var/data, so I am
>>> surprised that this is causing a spike on the root partition's disk usage.
>>> It seems like this is a bug with the go server temporarily using the
>>> server's partition rather than the artifact's partition as a store for
>>> uploaded artifacts. I am using gocd 17.11.
>>>
>>> Thanks
>>>
>>>
>>>
>>> On Monday, January 29, 2018 at 1:56:16 PM UTC, Edward Seager wrote:
>>>>
>>>> I've just had a similar notification which I am confused about:
>>>>
>>>> The email has been sent out automatically by the Go server at
>>>>> (10.6.42.142) to Go administrators.
>>>>> This server has less than 1024Mb of disk space available at /
>>>>> var/lib/go-server/db to store data. When the available space goes
>>>>> below 100Mb, Go will stop scheduling. Please ensure enough space is
>>>>> available.
>>>>
>>>>
>>>>
>>>> Looking at the usage, the root partiton has 4.2G free, so I would not
>>>> expect to see this error.
>>>>
>>>>
>>>> Filesystem             Size  Used Avail Use% Mounted on
>>>> udev                   7.9G     0  7.9G   0% /dev
>>>> tmpfs                  1.6G   25M  1.6G   2% /run
>>>> */dev/mapper/vg00-root   18G   13G  4.2G  75% /*
>>>> tmpfs                  7.9G     0  7.9G   0% /dev/shm
>>>> tmpfs                  5.0M     0  5.0M   0% /run/lock
>>>> tmpfs                  7.9G     0  7.9G   0% /sys/fs/cgroup
>>>> /dev/vda1              453M   55M  371M  13% /boot
>>>> /dev/mapper/vg00-data  493G  355G  113G  76% /var/data
>>>> tmpfs                  1.6G     0  1.6G   0% /run/user/1674
>>>> tmpfs                  1.6G     0  1.6G   0% /run/user/20240
>>>>
>>>>
>>>> Could anyone advise what the issue is here? Note, the server is
>>>> virtualised so it may be that the storage became inaccessibly briefly which
>>>> could have caused the error.
>>>>
>>>> Many thanks
>>>>
>>>>
>>>> On Friday, September 26, 2014 at 2:38:32 PM UTC+1, CI Rookie wrote:
>>>>>
>>>>> Got a notification email saying "This server has less than 1024Mb of
>>>>> disk space available at C:\Program Files (x86)\Go Server\db to store data.
>>>>> When the available space goes below 100Mb, Go will stop scheduling. Please
>>>>> ensure enough space is available"
>>>>>
>>>>> In which situation Go email this notification?..obviously my C drive
>>>>> has plenty of free space...not sure what triggered this warning..
>>>>>
>>>>> any idea?
>>>>>
>>>>>
>>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "go-cd" group.
>>> To unsubscribe from this group and stop receiving emails from it, send
>>> an email to [email protected].
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>
>> --
> You received this message because you are subscribed to the Google Groups
> "go-cd" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"go-cd" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to