Re: message type 0x49 arrived from server while idle
The solution David and others found is not applicable to my problem because I'm already running apache 1.3.27 (not 1.3.28 which seems to be causing this trouble) I'm not sure I understand what you're saying here. At the beginning of your message it looked like you were using 1.3.27, not 1.3.28. Are you saying that the problem is present with a 1.3.27 Apache but not with 1.3.28? If so then could you not simply stay with 1.3.27? Ged, thanks for your answer. I am currently using 1.3.27 (haven't tried 1.3.28) - David's problem went away after reverting from 1.3.28 to 1.3.27 but my problem persists even with 1.3.27 ... cheers, tobias --- http://www.funkreich.de // may the funk be with you
message type 0x49 arrived from server while idle
I have developed an application based on Apache 1.3.27, mod_perl 1.27, Postgres 7.3.2, Class::DBI 0.94 (utilizes Ima::DBI) on SuSE Linux 7.0 with a 2.4.20 kernel. I already posted my problem to the Class::DBI userlist but unfortunately didn't get any answers yet so I'm trying my luck here hoping that somebody can help. The problem described below is very similar to the one posted by David Wheeler (of bricolage fame) some time ago (Topic mod_perl + Apache 1.3.28, http://www.gossamer-threads.com/perl/mailarc/gforum.cgi?post=75733;search_string=0x49;guest=1213631t=search_engine#75733) Here we go: The problem is that everytime more than one request at a time is made to the server I find the following error message in my logs and from then on apache seems to be stalled and is unable to process any further requests: message type 0x49 arrived from server while idle Nobody seems to have seen this error message before (I grep'd through my perl lib dir and found nothing). It seems to have got something to do with many apache children accessing the postgres database (I'm using Apache::DBI but also tried without it) because the problems are completely gone when starting apache in single-process mode. I'm really stuck with this problem and have absolutely no clue where to look. I tried building some stripped-down scripts to reproduce the error in a smaller environment but strangely that always worked. I'd be very grateful if somebody here could share some ideas on this. The solution David and others found is not applicable to my problem because I'm already running apache 1.3.27 (not 1.3.28 which seems to be causing this trouble) Thank you very much in advance! -- Cheers, Tobias --- http://www.funkreich.de // may the funk be with you
Re: message type 0x49 arrived from server while idle
Hi there, On Fri, 19 Sep 2003, Tobias Kremer wrote: I have developed an application based on Apache 1.3.27, mod_perl 1.27, Postgres 7.3.2, Class::DBI 0.94 (utilizes Ima::DBI) on SuSE Linux 7.0 with a 2.4.20 kernel. The problem is that everytime more than one request at a time is made to the server I find the following error message in my logs and from then on apache seems to be stalled and is unable to process any further requests: message type 0x49 arrived from server while idle Nobody seems to have seen this error message before (I grep'd through my perl lib dir and found nothing). I think it's from Postgres, not Perl. Try Googling for message type arrived from server while idle It seems to have got something to do with many apache children accessing the postgres database I'm sure that's right, I think you're confusing Postgres somehow. There are suspicions about threads in some of the messages that Google found with the query above. I've never used Postgres so I wouldn't know where to point you, but have you checked the Postgres docs for that message? There are a few references such as this http://developer.postgresql.org/docs/pgsql/src/interfaces/libpq/fe-protocol3.c fairly late on in the Google results. The solution David and others found is not applicable to my problem because I'm already running apache 1.3.27 (not 1.3.28 which seems to be causing this trouble) I'm not sure I understand what you're saying here. At the beginning of your message it looked like you were using 1.3.27, not 1.3.28. Are you saying that the problem is present with a 1.3.27 Apache but not with 1.3.28? If so then could you not simply stay with 1.3.27? 73, Ged.
