0

We are facing an issue with send mail. Earlier mails were going through our system but we have found that send-mail is taking maximum system usage. After going through the log file we have found following error:

Jan 23 14:44:08 mss60 sm-mta[28538]: s0N9Cjgc028538: SYSERR(root): collect: Cannot write ./dfs0N9Cjgc028538 (sm_io_flush||sm_io_error, uid=0, gid=129)
Jan 23 14:44:08 mss60 sm-mta[28538]: s0N9Cjgc028538: from=<>, size=-2147483648, class=0, nrcpts=1, proto=ESMTP, daemon=MTA-v4, relay=localhost [127.0.0.1]
Jan 23 14:44:08 mss60 sm-msp-queue[25416]: s0K3U2cx005447: to=postmaster, delay=3+05:35:56, xdelay=00:01:23, mailer=relay, pri=-2126752017, relay=[127.0.0.1] [127.0.0.1], dsn=4.0.0, stat=Deferred: 421 4.3.0 collect: Cannot write ./dfs0N9Cjgc028538 (sm_io_flush||sm_io_error, uid=0, gid=129)
Jan 23 14:45:51 mss60 sm-mta[30145]: s0N9E8T8030145: SYSERR(root): collect: Cannot write ./dfs0N9E8T8030145 (sm_io_flush||sm_io_error, uid=0, gid=129)
Jan 23 14:45:51 mss60 sm-mta[30145]: s0N9E8T8030145: from=<>, size=-2147483648, class=0, nrcpts=1, proto=ESMTP, daemon=MTA-v4, relay=localhost [127.0.0.1]
Jan 23 14:45:51 mss60 sm-msp-queue[25416]: s0K2U1L2001178: to=postmaster, delay=3+06:35:19, xdelay=00:01:43, mailer=relay, pri=-2126482017, relay=[127.0.0.1] [127.0.0.1], dsn=4.0.0, stat=Deferred: 421 4.3.0 collect: Cannot write ./dfs0N9E8T8030145 (sm_io_flush||sm_io_error, uid=0, gid=129)
Jan 23 14:47:34 mss60 sm-mta[32226]: s0N9Fq89032226: SYSERR(root): collect: Cannot write ./dfs0N9Fq89032226 (sm_io_flush||sm_io_error, uid=0, gid=129)
Jan 23 14:47:34 mss60 sm-mta[32226]: s0N9Fq89032226: from=<>, size=-2147483648, class=0, nrcpts=1, proto=ESMTP, daemon=MTA-v4, relay=localhost [127.0.0.1]
Jan 23 14:47:34 mss60 sm-msp-queue[25416]: s0K1A1Ja006006: to=postmaster, delay=3+07:59:08, xdelay=00:01:42, mailer=relay, pri=-2126122017, relay=[127.0.0.1] [127.0.0.1], dsn=4.0.0, stat=Deferred: 421 4.3.0 collect: Cannot write ./dfs0N9Fq89032226 (sm_io_flush||sm_io_error, uid=0, gid=129)

After going through the above error, we anticipated that /var/spool/mqueue-client is having permission issue. so we have given 770 to mqueue-client. After that we have restarted the Apache but found that mail is not going. Any suggestion/solution in this regard would be highly appreciable.

Ethen
  • 11
  • 3
  • Please post the permissions of /etc/mail/submit.cf, /usr/libexec/sendmail/sendmail, /var/spool/clientmqueue and /var/spool/mqueue – Vanadis Jan 27 '14 at 08:42
  • `/etc/mail/submit.cf`=644, `/usr/libexec/sendmail/sendmail` (send mail is not here inside libexec, it is at `/usr/sbin/sendmail` and permission is 777), `/var/spool/clientmqueue`=770 and `/var/spool/mqueue`=740 – Ethen Jan 27 '14 at 09:37
  • 2
    Did you say that `/usr/sbin/sendmail` has 777 permission? With a security hole THAT big, the only thing I would do is to consider the machine compromised, [nuke it from orbit](http://serverfault.com/q/218005/126632) and rebuild it from scratch. You might even discover which bit of hardware is bad while doing so. – Michael Hampton Jan 27 '14 at 13:48
  • Can you check owner:group for `/var/spool` and sub files/folders ? – krisFR Jan 27 '14 at 14:33
  • Michael Hampton♦ , We have given that permission deliberately in order to make it work. Apart from this there is not security leaks and all we have verified it. – Ethen Jan 28 '14 at 07:01
  • @user2196728 it is smmsp – Ethen Jan 28 '14 at 07:24

0 Answers0