Ah, it seems you aren't affiliated with Google. I'll see about filing a bug
report or something myself then.
Anyway, thanks for the help =)
Julian
On Sunday, 25 August 2013 13:33:47 UTC+2, Julian Kent wrote:
>
> Thanks Vinny
>
> I've put a delay of 5 seconds in, and reduced the number of items
Thanks Vinny
I've put a delay of 5 seconds in, and reduced the number of items in each
set_multi_async from 32 to 20.
However, this is still a new issue, since I didn't have any problems like
this until 1.8.x, even without any delay and dumping as fast as I could.
Could you notify the dev guy
On Sat, Aug 24, 2013 at 9:50 AM, Julian Kent wrote:
> In the appstats, further down I'm seeing memcache.Set() entries with times
> such as "real=8831941ms", which is clearly where it is hanging. Personally,
> I don't consider 8000 seconds, that is over 2 hours, reasonable latency in
> a single me
OK, I managed to catch one. I saved the appstats page, as well as the logs,
here:
https://dl.dropboxusercontent.com/u/4066735/LiteRumble_error/LiteRumble_error.zip
If you want to step through code the latest is available here:
https://bitbucket.org/jkflying/literumble/src/3542f1e636fe8675cf6e5b701
On Fri, Aug 23, 2013 at 1:36 AM, Julian Kent wrote:
> Will do. Right now it is tough to catch these though, they only happen
> ~20% of the time. It always seems to happen while I'm asleep :-/ Is there
> some way of getting appstats to save to db so I can see it later?
>
20% occurrence rate is m
On Thursday, 8 August 2013 00:23:30 UTC+2, Vinny P wrote:
> What does the variable *botsdict* represent? Based on the naming and
> function calls, I assume that it's a dictionary, but what type of objects
> is it containing? Is it possible that the contained objects are doing some
> kind of p
On Tue, Aug 6, 2013 at 3:29 AM, Julian Kent wrote:
> Here is the relevant code between those two logging points:
>
> client = memcache.Client()
> if len(botsdict) > 0:
> splitlist = dict_split(botsdict,32)
>
What does the variable *botsdict* re
I have a batch processing Backend (B4) which does a bunch of
unpickle/pickle and Numpy/array stuff. Recently I noticed that I was
getting much higher backend charges, which would hit quota almost every
time, so I migrated to Modules (also B4), thinking that might solve it.
However, I still see t
