[ic] Interchange crashing when updating order status in admin.

[Mike Heins]

Quoting Cameron G (ritontor at icenet.com.au):
> Here's an interesting one, and I'm at a bit of a loss to explain it so far.
> Here's the overview.
> 
> When processing orders and updating their status on one of the sites that
> run on a particular server, IC crashes - most of the time, but not all of
> the time. The other sites on the machine work perfectly. Here's an example
> of strace -f -p output that shows where I *think* the child process is
> breaking, incriminating details (like the admin username, which is where the
> issue may be?) have been changed. Up until this point, both traces appear to
> be basically identical...
> 
> 
> THE BAD STRACE:
> 
> [pid  7000] <... read resumed>
> "\5\0\0\0\r\0\0\0\3510\0\0007\0\0\0\2620\0\0D\0\0\0D6\0"..., 4096) = 4096
> [pid  6999] <... rt_sigprocmask resumed> [], 8) = 0
> [pid  7000] lseek(12, 12288, SEEK_SET <unfinished ...>
> [pid  6999] rt_sigprocmask(SIG_BLOCK, NULL,  <unfinished ...>
> [pid  7000] <... lseek resumed> )       = 12288
> [pid  6999] <... rt_sigprocmask resumed> [], 8) = 0
> [pid  7000] read(12,  <unfinished ...>
> [pid  6999] rt_sigprocmask(SIG_BLOCK, NULL,  <unfinished ...>
> [pid  7000] <... read resumed> "cusername\tpassword\tname\tlast_log"...,
> 178) = 178

This looks like you are running your userdb on GDBM or some other non-SQL
database. Not recommended, especially for busy sites. Interchange does a lot
of database updates in the latest foundation catalogs, and if you have 
a lot of logged-in users this can easily happen.

It probably means you are getting caught up in write contention that can't
be resolved.

Try switching the userdb, transactions, and orderline tables to SQL and
see what happens then.

-- 
Mike Heins
Perusion -- Expert Interchange Consulting    http://www.perusion.com/
phone +1.513.523.7621      <mike at perusion.com>

Friends don't let friends use Outlook. -- Bob Blaylock