[ic] Random GPG failures

Dan Browning interchange-users@interchange.redhat.com
Fri May 31 13:01:01 2002


At 12:32 PM 5/31/2002 -0400, you wrote:
>Hello Everyone,
>
>OK OK this is a repost with some updates,  I found the thread titled 
>"Intermittent PGP failures & Lost CC info".  The only solution that people 
>seem to have come up with is "cron job using interchange -r"  I also ran 
>through my log files again, and found the PGP error (I had been searching 
>for gpg rather than PGP previously) I too get PGP failed with error 3702 
>90% of the times that it fails.
>
>I guess my question now is has anyone found truely what the problem is yet?
>
>
>
>
>         I remember a few months back a thread dealing with random GPG 
> encryption failures, I spent a while today trying to find it, but 
> couldn't, so I was hoping that perhaps we could rehash this...
>
>I have a RH7.1 with all the latest patches and updates....  Running on a 
>Dell PowerApp
>Hardware
>700MHZ celeron "Sigh..... I know, customer has been reluctant to upgrade"
>512MB Ram
>1GB swap
>20GB WD HD's Ext3 mirrored (Raid1)  hdparm gives roughly 21MB/sec IO
>
>Software
>kernel 2.4.9-31
>perl 5.6.1
>IC 4.8.4
>GNUPG 1.0.6.
>
>The site gets roughly around 100 orders a day, the problem is 
>approximately 3 out of 100 orders do not get encrypted, and instead 
>display the CC as 5436**********.  With no errors in the logs that I can 
>find...  the /tmp dir has a few 0 length pgp.xxxxxx.out files.
>
>If I could get some more info regarding this problem, and it's appropriate 
>solution I would be gratefull....
>
>
>
>Brian Kosick
>Web Programmer
>New Age Consulting Service, Inc.
>216-619-2000
>briank@nacs.net

I think it was determined that Perl signal handling is the problem and the 
solution is to switch to RPC mode with MaxServers 0.


+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| Dan Browning, Kavod Technologies <db@kavod.com>
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
A CONS is an object which cares.
                 -- Bernie Greenberg.