Gamera Dead
Gamera Dead
Scott;
This is the second or third time this has happend, 3 different builds of redhat, rh9, fc2 and fc3. everything is happily chomping away at spam.... then one day it just stops accepting connections. ie its "listening" but refusing all connections even after restarting server, i can see qmail/clam/spama all running no load issue nothing in maillog accept clamav trying to update its db. any ideas? Before i redo system.
This is the second or third time this has happend, 3 different builds of redhat, rh9, fc2 and fc3. everything is happily chomping away at spam.... then one day it just stops accepting connections. ie its "listening" but refusing all connections even after restarting server, i can see qmail/clam/spama all running no load issue nothing in maillog accept clamav trying to update its db. any ideas? Before i redo system.
-
- Atomicorp Staff - Site Admin
- Posts: 8355
- Joined: Wed Dec 31, 1969 8:00 pm
- Location: earth
- Contact:
I'm still working on it, but I've got a big update to PG in atomic-testing (3es.atomicrocketturtle.com/atomic-testing/gamera/$releasever) if you want to see if that makes a difference. You'll need to add in Dag's archive (dag.wieers.com) to pull some components for this update. Use: yum install project-gamera
Scott;
Thanks for you time, i just wanted to post some revelant maillog also for others to consider, its very strange. after the last message it hasnt worked since. very strange.
ay 18 16:54:14 gamera qmail-scanner[21167]: SA:SPAM-DELETE:RC:0(24.232.90.250):SA:1(6.6/5.0): 0 1310 tamera_shaver@bac-cpa.com changedtoprotect@fastq.com been_trying_to_contact_you_the_past_few_d
ays <200341361351.14342.aseptic@debutante.peternixon.net> gamera.fastq.com111646044868021167-unpacked:1310
May 18 16:54:14 gamera spamd[18837]: connection from gamera.fastq.com [127.0.0.1] at port 47043
May 18 16:54:14 gamera spamd[18837]: checking message (unknown) for qmailq:104.
May 18 16:54:14 gamera spamd[15094]: server hit by SIGCHLD
May 18 16:54:14 gamera spamd[15094]: handled cleanup of child pid 13267
May 18 16:54:14 gamera spamd[15094]: server successfully spawned child process, pid 21228
May 18 16:54:14 gamera spamd[15094]: server hit by SIGCHLD
May 18 16:54:14 gamera spamd[15094]: handled cleanup of child pid 18837
May 18 16:54:14 gamera spamd[15094]: server successfully spawned child process, pid 21229
May 18 16:54:15 gamera spamd[15094]: server hit by SIGCHLD
May 18 16:54:15 gamera spamd[15094]: handled cleanup of child pid 21229
May 18 16:54:15 gamera spamd[15094]: server successfully spawned child process, pid 21274
May 18 16:54:15 gamera spamd[15094]: server hit by SIGCHLD
May 18 16:54:15 gamera spamd[15094]: handled cleanup of child pid 21228
May 18 16:54:15 gamera spamd[15094]: server successfully spawned child process, pid 21275
May 18 16:54:15 gamera spamd[15094]: server hit by SIGCHLD
May 18 16:54:15 gamera spamd[15094]: handled cleanup of child pid 17819
May 18 16:54:16 gamera spamd[15094]: server successfully spawned child process, pid 21276
May 18 16:54:16 gamera spamd[15094]: server hit by SIGCHLD
May 18 16:54:16 gamera spamd[15094]: handled cleanup of child pid 14082
May 18 16:54:16 gamera spamd[15094]: server killed by SIGTERM, shutting down
May 18 16:54:15 gamera freshclam[11994]: Received signal 15, terminating
May 18 16:54:16 gamera clamd[15192]: Socket file removed.
Thanks for you time, i just wanted to post some revelant maillog also for others to consider, its very strange. after the last message it hasnt worked since. very strange.
ay 18 16:54:14 gamera qmail-scanner[21167]: SA:SPAM-DELETE:RC:0(24.232.90.250):SA:1(6.6/5.0): 0 1310 tamera_shaver@bac-cpa.com changedtoprotect@fastq.com been_trying_to_contact_you_the_past_few_d
ays <200341361351.14342.aseptic@debutante.peternixon.net> gamera.fastq.com111646044868021167-unpacked:1310
May 18 16:54:14 gamera spamd[18837]: connection from gamera.fastq.com [127.0.0.1] at port 47043
May 18 16:54:14 gamera spamd[18837]: checking message (unknown) for qmailq:104.
May 18 16:54:14 gamera spamd[15094]: server hit by SIGCHLD
May 18 16:54:14 gamera spamd[15094]: handled cleanup of child pid 13267
May 18 16:54:14 gamera spamd[15094]: server successfully spawned child process, pid 21228
May 18 16:54:14 gamera spamd[15094]: server hit by SIGCHLD
May 18 16:54:14 gamera spamd[15094]: handled cleanup of child pid 18837
May 18 16:54:14 gamera spamd[15094]: server successfully spawned child process, pid 21229
May 18 16:54:15 gamera spamd[15094]: server hit by SIGCHLD
May 18 16:54:15 gamera spamd[15094]: handled cleanup of child pid 21229
May 18 16:54:15 gamera spamd[15094]: server successfully spawned child process, pid 21274
May 18 16:54:15 gamera spamd[15094]: server hit by SIGCHLD
May 18 16:54:15 gamera spamd[15094]: handled cleanup of child pid 21228
May 18 16:54:15 gamera spamd[15094]: server successfully spawned child process, pid 21275
May 18 16:54:15 gamera spamd[15094]: server hit by SIGCHLD
May 18 16:54:15 gamera spamd[15094]: handled cleanup of child pid 17819
May 18 16:54:16 gamera spamd[15094]: server successfully spawned child process, pid 21276
May 18 16:54:16 gamera spamd[15094]: server hit by SIGCHLD
May 18 16:54:16 gamera spamd[15094]: handled cleanup of child pid 14082
May 18 16:54:16 gamera spamd[15094]: server killed by SIGTERM, shutting down
May 18 16:54:15 gamera freshclam[11994]: Received signal 15, terminating
May 18 16:54:16 gamera clamd[15192]: Socket file removed.
Scott;
other than running yum update about a month ago no, the only modification or addition i did was some statistics program for spamstats. but i did that shortly after the box was created in Novemberish, believe it was called spamstats.pl ..... other than that it was pretty vanilla. oh also /etc/tcpcontrol/smtp.rules gets modified all kinds of cidrs that are blocked.
i have tried clearing that file to just the 1 default line. still get server is busy from email program.
other than running yum update about a month ago no, the only modification or addition i did was some statistics program for spamstats. but i did that shortly after the box was created in Novemberish, believe it was called spamstats.pl ..... other than that it was pretty vanilla. oh also /etc/tcpcontrol/smtp.rules gets modified all kinds of cidrs that are blocked.
i have tried clearing that file to just the 1 default line. still get server is busy from email program.
Scott;
For grins i just ran yum update gamera and recieved;
Gathering header information file(s) from server(s)
Server: Red Hat Linux 9 - i386 - Base
Server: Atomic Rocket Turtle - 9 - Project Gamera
retrygrab() failed for:
http://shinn.net/pub/Linux/atomic/gamer ... eader.info
Executing failover method
failover: out of servers to try
Error getting file http://shinn.net/pub/Linux/atomic/gamer ... eader.info
[Errno 4] IOError: HTTP Error 404: Not Found
<root@gamera:~>
For grins i just ran yum update gamera and recieved;
Gathering header information file(s) from server(s)
Server: Red Hat Linux 9 - i386 - Base
Server: Atomic Rocket Turtle - 9 - Project Gamera
retrygrab() failed for:
http://shinn.net/pub/Linux/atomic/gamer ... eader.info
Executing failover method
failover: out of servers to try
Error getting file http://shinn.net/pub/Linux/atomic/gamer ... eader.info
[Errno 4] IOError: HTTP Error 404: Not Found
<root@gamera:~>
Scott;
Your not supposed to tell me that! ..... gamera "ours" is running atomic-testing. So far other than those 2 issues i pointed out the clam.conf user and gpg thing all seems to be marrily chomping away at spam. in fact using a score of 5 since about 5pm mst it has deleted 21,940 messages
and 301 virus's.
However i had noticed you arent using razor or dcc at least not at this time.
so far so good!
Your not supposed to tell me that! ..... gamera "ours" is running atomic-testing. So far other than those 2 issues i pointed out the clam.conf user and gpg thing all seems to be marrily chomping away at spam. in fact using a score of 5 since about 5pm mst it has deleted 21,940 messages
and 301 virus's.
However i had noticed you arent using razor or dcc at least not at this time.
so far so good!

-
- Atomicorp Staff - Site Admin
- Posts: 8355
- Joined: Wed Dec 31, 1969 8:00 pm
- Location: earth
- Contact:
Hence the term "atomic-testing", you're literally in the middle of development cycle. It works like this:
1) I make changes to the package, compile, build, etc
2) publish to atomic-testing
3) yum install/update the packages
4) test features, installation, etc
5) goto 1 until everything works like I think it should
If its in atomic-testing, its absolutely not feature complete. Except in very rare cases, I dont install everything outside of an rpm, get it all worked out and then rpm it up. As I'm using rpms as part of the development process, atomic-testing is what you get before even I install/test those packages most of the time.
Another thing to watch out for, a lot of the time given the number of revisions I'll make to a package, I don't update the revision number (the -3 -4 -5 type stuff), so the only way you'll know if you have the latest is by looking at the timestamps and the size of the file. Definitely keep that in mind if you're running it on a non-test server, since you could end up with something that says its version 1.0, but in reality its a few revisions older.
1) I make changes to the package, compile, build, etc
2) publish to atomic-testing
3) yum install/update the packages
4) test features, installation, etc
5) goto 1 until everything works like I think it should
If its in atomic-testing, its absolutely not feature complete. Except in very rare cases, I dont install everything outside of an rpm, get it all worked out and then rpm it up. As I'm using rpms as part of the development process, atomic-testing is what you get before even I install/test those packages most of the time.
Another thing to watch out for, a lot of the time given the number of revisions I'll make to a package, I don't update the revision number (the -3 -4 -5 type stuff), so the only way you'll know if you have the latest is by looking at the timestamps and the size of the file. Definitely keep that in mind if you're running it on a non-test server, since you could end up with something that says its version 1.0, but in reality its a few revisions older.