This morning I discovered this in the logwatch report for our external MX backup host.
STARTTLS: write error=syscall error (-1), errno=32, get_error=error:00000000:lib(0):func(0):reason(0), retry=99, ssl_err=5: 206 Time(s)
I also see many entries similar to this:
8: fl=0x802, mode=140777: SOCK inet04.mississauga.harte-lyne.ca/34091->(Transport endpoint is not connected): 1 Time(s) MCI@0x8055b34: flags=27c86c<CACHED,ESMTP,SIZE,8BITMIME,DSN,INMIME,AUTH,AUTHACT,ENHSTAT,PIPELINED,TLSACT>, errno=32, herrno=0, exitstat=75, state=8, pid=0, maxsize=20480000, phase=client DATA 354, mailer=esmtp, status=4.4.2, rstatus=(null), host=inet08.hamilton.harte-lyne.ca., lastuse=Mon Nov 24 10:56:35 2014\n: 1 Time(s) MCI@0x8053aa4: flags=27c86c<CACHED,ESMTP,SIZE,8BITMIME,DSN,INMIME,AUTH,AUTHACT,ENHSTAT,PIPELINED,TLSACT>, errno=32, herrno=0, exitstat=75, state=8, pid=0, maxsize=20480000, phase=client DATA 354, mailer=esmtp, status=4.4.2, rstatus=(null), host=inet08.hamilton.harte-lyne.ca., lastuse=Mon Nov 24 18:41:32 2014\n: 1 Time(s) MCI@0x8053aa4: flags=27c86c<CACHED,ESMTP,SIZE,8BITMIME,DSN,INMIME,AUTH,AUTHACT,ENHSTAT,PIPELINED,TLSACT>, errno=32, herrno=0, exitstat=75, state=8, pid=0, maxsize=20480000, phase=client DATA 354, mailer=esmtp, status=4.4.2, rstatus=(null), host=inet08.hamilton.harte-lyne.ca., lastuse=Tue Nov 25 07:41:32 2014\n: 1 Time(s)
I have not seen this sort of thing before and I am wondering what it means? Googling has not provided anything useful. In case this was an incipient HDD error I rebooted with /forcefsck but on CentOS-5 one obtains no log of the rc report. I am also not entirely clear on whether or not /forcefsck actually repairs any errors. Thus it may not mean much to have done this.