3:08 p.m.
Forgive me if this is OT. For the past few days my IMAP connection has been
dropping every twenty seconds or so. This is what .muttdebug0 ends with:
[2014-03-05 11:43:29] 4> a0083 NOOP
[2014-03-05 11:43:29] SSL error: error:0B07C065:x509 certificate
routines:X509_STORE_add_cert:cert already in hash table
[2014-03-05 11:43:29] Connection to [mail.domain.tld] closed
[2014-03-05 11:43:31] imap_cmd_step: Error reading server response.
[2014-03-05 11:43:31] Mailbox closed
I doubt it’s related to the recent GnuTLS bug because I compiled with OpenSSL
instead. In case it’s any use:
Compile options:
-DOMAIN
+DEBUG
+HOMESPOOL -USE_SETGID +USE_DOTLOCK -DL_STANDALONE +USE_FCNTL -USE_FLOCK
-USE_POP +USE_IMAP -USE_SMTP
+USE_SSL_OPENSSL -USE_SSL_GNUTLS +USE_SASL +USE_GSS +HAVE_GETADDRINFO
-HAVE_REGCOMP +USE_GNU_REGEX
+HAVE_COLOR +HAVE_START_COLOR +HAVE_TYPEAHEAD +HAVE_BKGDSET
+HAVE_CURS_SET +HAVE_META +HAVE_RESIZETERM
+CRYPT_BACKEND_CLASSIC_PGP +CRYPT_BACKEND_CLASSIC_SMIME +CRYPT_BACKEND_GPGME
+EXACT_ADDRESS -SUN_ATTACHMENT
+ENABLE_NLS -LOCALES_HACK +HAVE_WC_FUNCS +HAVE_LANGINFO_CODESET
+HAVE_LANGINFO_YESEXPR
+HAVE_ICONV -ICONV_NONTRANS +HAVE_LIBIDN +HAVE_GETSID +USE_HCACHE +NOTMUCH
-MIXMASTER
Is this even a mutt problem? I can’t think of anything relevant that changed lately. My
system is
Darwin [hostname] 13.1.0 Darwin Kernel Version 13.1.0: Thu Jan 16 19:40:37 PST 2014;
root:xnu-2422.90.20~2/RELEASE_X86_64 x86_64
Thanks.
2:41 p.m.
On further investigation, there appears to be a server-side component to the
problem, as it happens exactly the same way on a different build, on a
different OS, on a different machine, in a different network, on the same
subnet as the mailserver. Unfortunately I don’t have shell access to the
mailserver.
I worked around the NOOP by explicitly setting imap_keepalive; the default is
supposed to be 900 seconds but apparently it’s much shorter.
Now I get the cert error and disconnection (and sometimes a segfault) when
changing mailboxes after too long - the trigger appears to be “Draining IMAP
command pipeline.”
Again, apologies if this is the wrong place for this inquiry.
2:58 p.m.
On Thu, 06 Mar 2014, Ludwig wrote:
I worked around the NOOP by explicitly setting imap_keepalive; the
default is
supposed to be 900 seconds but apparently it’s much shorter.
Addendum: I’ve set imap_keepalive to 1800 seconds but now the NOOP gets sent
at exactly 600, which is better than the 40 or so that it was doing before, but
still, this may be a bug.
3704
days inactive
3705
days old
mutt-kz@lists.fedoraproject.org
2 comments
1 participants
participants (1)
-
Ludwig