efa-project.org

On our production EFA, the daily quarantine report has been broken for some time, we have updated to 3.0.0.6 and I manually ran the quarantine_report.php, it said creating report for each user and e-mailing but it was never received by any of the users. This morning after updating to 3.0.0.6 last night, mailscanner appeared to hang scanning the same messages over and over again, up to 80,000 showing processed but only 800 showed in quarantine, rebooted EFA and it was fine after that, but was just very weird and had read that it was a previous issue that should've been resolved with 3.0.0.6. Also, on my home lab EFA the daily quarantines are coming through but now are showing every e-mail scanned instead of just spam. Thanks in advance, man.

It happened again last night, looks like at 4:00am. I am assuming there's a cronjob that's causing mailscanner to crash...afterwards it passes mail very slowly but all unscanned/clean until reboot. Please assist, thank you!

Is your localhost whitelisted?

You should have an entry in your whitelist....

From: 127.0.0.1
To: default

As for the quarantine report...

/viewtopic.php?f=14&t=801

Yeah whitelist is present. Most pressing issue is whatever happens at 4am causing spamassassin to crash then our inbound queue backs up and loops on the same emails showing # processed in the tens of thousands. Any ideas?

In /etc/cron.daily...

Those are what execute around that time.

I would recommend executing each one, one at a time, to see which one is triggering the problem...once you know which cron is causing the issue, post back the results.

I saw in another post about a similar issue where darky mentioned maybe needing more resources so I gave VM 2gb more ram.

viewtopic.php?f=13&t=818
This is post I read... Will check cron.daily in a lil bit and let you know.

I wasn't able to run the cron.daily jobs but it didn't crash last night after having given the VM more resources. I bumped it up from 2gb to 4gb RAM and all was well this morning. I will watch it for a couple days and see how it goes. But on this same box our daily reports are not sending. localhost is whitelisted, and job appears to run properly.

You just gave me an aha! moment....

I think I know why the processing attempts database wasn't working properly...

As for your reports not sending...

What happens when you run the job yourself? Does it list all the recipients to the console?



?

when i manually run the job it goes through each user one by one saying creating report then e-mailing but the report never arrives...

Well, that means they are being generated, at least...

What appears in the maillog..../var/log/maillog during this time?

As a test just try setting the Postfix General Settings option "What domains to receive mail for" to "Whole domain" and restart postfix. But take a copy of the contents of the box first as you will need that to set it back again.

Same issue here. Nothing in the maillog about it being sent. But the reports look like they generate after running quarantine_report.php

Actually troubleshooting it more, here is output from maillog when php executes:

Jan 22 10:33:00 shield postfix/smtpd[23207]: connect from localhost[127.0.0.1]
Jan 22 10:33:00 shield postfix/smtpd[23207]: match_hostname: localhost ~? 127.0. 0.0/8
Jan 22 10:33:00 shield postfix/smtpd[23207]: match_hostaddr: 127.0.0.1 ~? 127.0. 0.0/8
Jan 22 10:33:00 shield postfix/smtpd[23207]: > localhost[127.0.0.1]: 220 shield. fortresst.com ESMTP Postfix
Jan 22 10:33:00 shield postfix/smtpd[23207]: xsasl_cyrus_server_create: SASL ser vice=smtp, realm=(null)
Jan 22 10:33:00 shield postfix/smtpd[23207]: name_mask: noanonymous
Jan 22 10:33:00 shield postfix/smtpd[23207]: watchdog_pat: 0x7f8490659220
Jan 22 10:33:00 shield postfix/smtpd[23207]: vstream_fflush_some: fd 9 flush 40
Jan 22 10:33:00 shield postfix/smtpd[23207]: vstream_buf_get_ready: fd 9 got 16
Jan 22 10:33:00 shield postfix/smtpd[23207]: < localhost[127.0.0.1]: EHLO localh ost
Jan 22 10:33:00 shield postfix/smtpd[23207]: > localhost[127.0.0.1]: 250-shield. fortresst.com
Jan 22 10:33:00 shield postfix/smtpd[23207]: > localhost[127.0.0.1]: 250-PIPELIN ING
Jan 22 10:33:00 shield postfix/smtpd[23207]: > localhost[127.0.0.1]: 250-SIZE 10 240000
Jan 22 10:33:00 shield postfix/smtpd[23207]: > localhost[127.0.0.1]: 250-ETRN
Jan 22 10:33:00 shield postfix/smtpd[23207]: > localhost[127.0.0.1]: 250-STARTTL S
Jan 22 10:33:00 shield postfix/smtpd[23207]: > localhost[127.0.0.1]: 250-AUTH CR AM-MD5 PLAIN DIGEST-MD5 LOGIN
Jan 22 10:33:00 shield postfix/smtpd[23207]: match_list_match: localhost: no mat ch
Jan 22 10:33:00 shield postfix/smtpd[23207]: match_list_match: 127.0.0.1: no mat ch
Jan 22 10:33:00 shield postfix/smtpd[23207]: > localhost[127.0.0.1]: 250-AUTH=CR AM-MD5 PLAIN DIGEST-MD5 LOGIN
Jan 22 10:33:00 shield postfix/smtpd[23207]: > localhost[127.0.0.1]: 250-ENHANCE DSTATUSCODES
Jan 22 10:33:00 shield postfix/smtpd[23207]: > localhost[127.0.0.1]: 250-8BITMIM E
Jan 22 10:33:00 shield postfix/smtpd[23207]: > localhost[127.0.0.1]: 250 DSN
Jan 22 10:33:00 shield postfix/smtpd[23207]: watchdog_pat: 0x7f8490659220
Jan 22 10:33:00 shield postfix/smtpd[23207]: vstream_fflush_some: fd 9 flush 217
Jan 22 10:33:00 shield postfix/smtpd[23207]: vstream_buf_get_ready: fd 9 got 6
Jan 22 10:33:00 shield postfix/smtpd[23207]: < localhost[127.0.0.1]: RSET
Jan 22 10:33:00 shield postfix/smtpd[23207]: > localhost[127.0.0.1]: 250 2.0.0 O k
Jan 22 10:33:00 shield postfix/smtpd[23207]: watchdog_pat: 0x7f8490659220
Jan 22 10:33:00 shield postfix/smtpd[23207]: vstream_fflush_some: fd 9 flush 14
Jan 22 10:33:00 shield postfix/smtpd[23207]: vstream_buf_get_ready: fd 9 got 6
Jan 22 10:33:00 shield postfix/smtpd[23207]: < localhost[127.0.0.1]: QUIT
Jan 22 10:33:00 shield postfix/smtpd[23207]: > localhost[127.0.0.1]: 221 2.0.0 B ye
Jan 22 10:33:00 shield postfix/smtpd[23207]: match_hostname: localhost ~? 127.0. 0.0/8
Jan 22 10:33:00 shield postfix/smtpd[23207]: match_hostaddr: 127.0.0.1 ~? 127.0. 0.0/8
Jan 22 10:33:00 shield postfix/smtpd[23207]: disconnect from localhost[127.0.0.1

My bad man! I messed up something in the images directory under html (a linked logo) and it was stopping the report, confirmed. something to do with $hdrs variable.