Re: [Sqlgrey-users] 1.6.x and 1.7.x freezes

[Ed S. (s. list) <sq...@cr...>]

Here's an update since a few things have changed on my server. It may be
too soon to see if that wondrous lockup bug has been resolved, but it
seems that it *MAY* have indeed been squashed.

Previously, as you know all too well, it would fail/lock-up within a
day's time.  Until the power outage, it was working flawlessly since the
upgrade to 1.6.4.

So, below is an updated "checklist" for your perusal.



Ed Shornock (sqlgrey list) wrote:
> Lionel Bouton wrote:

>>Do you have freeze problems: (yes/no)
> 
> yes  :(

The answer is now "maybe"

With Sqlgrey 1.6.4 (installed on July 8th) and the following settings:
---start---
$ grep -v '#' /etc/sqlgrey/sqlgrey.conf  |strings
conf_dir = /etc/sqlgrey
loglevel = 4
user = sqlgrey
group = nogroup
db_type = Pg
db_name = sqlgrey
db_host = localhost
db_user = sqlgrey
db_pass = ANotSoCommonPassword
admin_mail = pos...@cr...
---end---

Sqlgrey was working *perfectly*.  Was, in the past tense, because a
f%#$#@ing brown-out made my "server" lose power. :(

So Sqlgrey stopped working through no fault of its own. :P


>>Latest SQLgrey version used: (1.6.[012] / 1.7.0)
1.6.4

>>OS/Distribution: (uname -a)
> 
Debian Sid (unstable)
$ uname -a
Linux darkside 2.6.12.2-p4 #1 Thu Jul 7 23:50:52 EDT 2005 i686 GNU/Linux

>>Perl version: (see perl -v)
$ perl -v

This is perl, v5.8.7 built for i386-linux-thread-multi




Note:  Recently, Perl packages were updated in Debian unstable with the
following ChangeLog:

perl (5.8.7-4) unstable; urgency=low

  * Don't generate broken md5sums for libperl5.8 (closes: #304640).
  * Build with gcc 4.0; restore standard (-O2) optimisation on arm, ia64
    and powerpc.  Remove gcc-2.95 build work-around for m68k.
  * Fix test of reenterant function return values which was causing
    perl to malloc itself to death if ERANGE was encountered before
    ENOENT (such as a long line in /etc/group; closes: #227621).

 -- Brendan O'Dea <bo...@de...>  Sat,  9 Jul 2005 11:14:13 +1000


About the time that it started locking up, the Perl ChangeLog showed:
perl (5.8.7-2) unstable; urgency=low

  * Sarge has released.  Rebuild for unstable.

 -- Brendan O'Dea <bo...@de...>  Tue,  7 Jun 2005 09:31:16 +1000



Is it related as a possible cause?  I have no idea.

>>Net-Server-Multiplex version: (perl -MNet::Server::Multiplex -e 'print
>>$Net::Server::Multiplex::VERSION')

> $ perl -MNet::Server::Multiplex -e 'print $Net::Server::Multiplex::VERSION'
> 0.88

>>How many mails/day:
Currently about 900/day