On 08.03.2012 11:25, Daniel Shahaf wrote:
> Stefan Fuhrmann wrote on Thu, Mar 08, 2012 at 07:33:03 +0100:
>> On 08.03.2012 02:13, Bert Huijben wrote:
>>>> -----Original Message-----
>>>> From: Stefan Sperling [mailto:stsp_at_elego.de]
>>>> Sent: donderdag 8 maart 2012 1:28
>>>> To: Stefan Fuhrmann
>>>> Cc: Subversion Development
>>>> Subject: Re: Revprop caching 'n stuff
>>>>
>>>> On Wed, Mar 07, 2012 at 11:40:40PM +0100, Stefan Fuhrmann wrote:
>>>>> Hi all,
>>>> And this shared mem approach will work with both threaded as well as
>>>> multi-process servers?
>>>>
>>>> If so, this sounds like a good plan to me.
>>> This still assumes that the repository is only used by a single machine,
>>> while using NFS paths with multiple servers is not uncommon. (I heard from a
>>> few of those setups on users_at_s.a.o)
>> That is correct. Those users will not be able to use
>> revprop caching. And this is one of the reasons why
>> it will need to be activated explicitly.
> Could those users use memcached for the revision proplists?
>
In theory, we could separate the revprop cache
and place it into a *single* memcached instance.
If all servers did that, it could be made to work.
However, our current use of memcached is
different as we allow for multiple instances to
used and there is a severe bug in APR-UTILS
that basically renders them useless:
http://svn.haxx.se/dev/archive-2011-07/0467.shtml
-- Stefan^2.
Received on 2012-03-09 10:48:01 CET