wiki outage

The wiki is having an outage, we're working on it.  Should
be up soon.  
We suspect our issues are because of mod_python.  The moin
developers 
suggest not using it.  So we'll go back to cgi (slow) and
deal with it 
(stable).

    -Mike

-- 
Fedora-websites-list mailing list
Fedora-websites-listredhat.com
https://www.redhat.com/mailman/listinfo/fedora-websi
tes-list

Mike McGrath wrote:
> The wiki is having an outage, we're working on it. 
Should be up 
> soon.  We suspect our issues are because of mod_python.
 The moin 
> developers suggest not using it.  So we'll go back to
cgi (slow) and 
> deal with it (stable).

Ok, so that experiment lasted almost 5 minutes   The box
crapped out 
just after it hit a load around 200.  We had a whole team of
sysadmins 
in #fedora-admin and did some things including enabling keep
alive (no 
one even remembers why it was off) but also lowering the
number of 
MaxRequestsPerChild significantly.  We were also working
with the Moin 
developers to get this going. 

The moin developers suggested using mod_fastCGI, but it has
an 
unfriendly license at present.   So we're back to mod_python
for the 
moment (other things were suggested and will be tested if
we're still 
seeing issues).  Things seem to be going well though, load
is low,  the 
box doesn't seem to be swapping much.  We'll have to keep an
eye on it 
and see how everything goes.

Anyone interested in any of the tests we ran and things we
changed let 
me know.  We're just going to let it be for a while and if,
in fact, we 
have fixed the issue then we'll migrate that back to the
other app 
servers (its currently going proxy -> app server no
balancing)

    -Mike

-- 
Fedora-websites-list mailing list
Fedora-websites-listredhat.com
https://www.redhat.com/mailman/listinfo/fedora-websi
tes-list