Mailing List Archive

Our bric_queued grows until it consumes about 60% of the server's total
memory before it "freezes up" and nothing further in the queue will ever
be executed. I simply keep an eye on it and manually restart
bric_queued whenever it exceeds about 20% memory consumption. Certainly
not a good solution - so if anyone has a better idea please let us know.

I had assumed this is caused a memory leak in or around Net::SSH2. Can
you confirm otherwise?

-Nick

On 8/16/2010 3:37 PM, Phillip Smith wrote:
> Just wondering if anyone who is using bric_queued in production is seeing the bric_queued processes growing in size to several gigabytes of memory? If so, what do you use to manage those processes and keep them from getting out of control?
>
> Phillip.
>
> --
> Phillip Smith // Simplifier of Technology // COMMUNITY BANDWIDTH
> www.communitybandwidth.ca // www.phillipadsmith.com
>
>
>
>
>
>
>
>

--
Nick Legg
Information Technology Services
Web Developer
Denison University
leggn@denison.edu

On Aug 16, 2010, at 12:37 PM, Phillip Smith wrote:

>
> Just wondering if anyone who is using bric_queued in production is seeing the bric_queued processes growing in size to several gigabytes of memory? If so, what do you use to manage those processes and keep them from getting out of control?

I committed some changes for 2.1 to address this issue:

http://github.com/bricoleurs/bricolage/commits/master

See especially all but the first commit on 2010-07-07 and the one commit on 2010-07-08.

Best,

David