IMAP problems

[kluss0]

After upgrading to Scalix 10 (enterprise), I am having problems with some peoples mailboxes being accessed via IMAP. Outlook with Scalix Connect works fine. SWA does not work at all (for anyone) due to SOAP errors. The IMAP error that I get (from Outlook 2003) is "Failed to update Headers. The TCP/IP connection was unexpectedly terminated by the server". On Evolution, I get "Error Refreashing Folder: Server response ended too soon". Thunderbird just stops responding with no error. On the server side, in /var/opt/scalix/logs/fatal I get:

Code: Select all

SERIOUS ERROR           IMAP Server Da(IMAP Server Pr) Tue Feb 21 07:39:57 2006
[OM 10270] Process about to terminate due to error.
Signal (Segmentation Violation) trapped by process 16173
Procedure trace follows:
  -> imapMimeGetStructure
  -> imapMimeOpenTempFile
  <- imapMimeOpenTempFile
  -> ct_FindDITEntryByRef
  -> ct_DecodeDirectRef
  <- ct_DecodeDirectRef
  -> ct_FindDITEntry
  <- ct_FindDITEntry
  <- ct_FindDITEntryByRef
  -> msl_readContainerOpenedParent
  -> ct_ReadContentRec 3001 0
  <- ct_ReadContentRec 3001 0
  -> msl_reload
  -> ct_ReadExtension 3001 len=2168 rec=3120
  <- ct_ReadExtension
  <- ct_ReadExtension 3001 len=2168 rec=3120
User Name: Kenneth E. Lussier / postal, aylus/CN=Kenneth Lussier
Pid of logging process: 16173


SERIOUS ERROR           IMAP Server Da(IMAP Server Pr) Tue Feb 21 07:39:57 2006
[OM 10272] BACKTRACE:
        /opt/scalix/lib/libom_er.so(er_add_backtrace+0xc6)[0x819f06]
        /opt/scalix/lib/libom_er.so[0x81a1d6]
        /opt/scalix/lib/libom_er.so(er_DumpProcAndExit+0x1f)[0x81a37f]
        /lib/tls/libpthread.so.0[0xa838b8]
        /opt/scalix/lib/libom_msl.so[0x37c778]
        /opt/scalix/lib/libom_msl.so[0x37c70f]
        /opt/scalix/lib/libom_msl.so(msl_reload+0xa80)[0x37d200]
        /opt/scalix/lib/libom_msl.so(msl_readContainerOpenedParent+0x857)[0x373b27]
        in.imap41d[0x80620a5]
        in.imap41d[0x805a9bb]
        in.imap41d[0x805ef29]
        in.imap41d[0x80602cd]
        in.imap41d[0x8060b86]
        /lib/tls/libc.so.6(__libc_start_main+0xe3)[0x1dde33]
        in.imap41d[0x804d981]
        User Name: Kenneth E. Lussier / postal, aylus/CN=Kenneth Lussier
        Pid of logging process: 16173

My server environment is: RHEL4 (ES), Scalix 10 Enterprise. This problem is only effecting certain users, and it can be duplicated using any IMAP client on both Windows and Linux. It also does not matter if I connect using SSL (stunnel) or standard IMAP. I have tried to run omtidyu but that did not change anything. Any suggestions?

Thanks,
Kenny

[markd]

It looks like the cached MIME header information has been corrupted. The easiest way to rebuild this for all your users is to do the following:

Add the following line to ~/sys/general.cfg
MSLDATA_SUBVERSION=1

restart the Mime Browser Controller, Local Delivery and IMAP
omoff -d 0 -w mime ld imap
omon -w mime ld imap

Then run
omtidyallu -M

This will take some time and slow all IMAP/SWA clients (the MIME version of the messages will be regenerated). But when it finishes, all should be well again.

You may want to try this on a single user first, in which case do the setting in general.cfg, restart services, then run
omtidyu -B -u user -M
then remove general.cfg setting (while the setting is there any use of IMAP/SWA will cause cached data to be updated for any user, so access will be slow until tidy is run)
You should then be able to access this user with imap correctly

[jch]

My first reaction here would be to run omscan against the user(s) with problems to see if there's some corruption in the message store. Try

Code: Select all

omsan -Avx -U "Kenneth E. Lussier"

(the name is cut'n'pasted from the error message).

Where we go next depends on what errors there are: there are basically two possibilities one is to run omscan to fix the problems and the other is to force the MSL data to be regenerated.

jch

[kluss0]

I made the change in general.cfg and ran omtidyu on my mailbox. It worked like a charm. I'll have to do it for everyone else. The interesting thing is that it seems to have fixed some of my SWA problems as well.

Thanks,
Kenny

[kluss0]

I managed to corrupt my mailbox again. I believe that I have found the culprit: Scalix Connect for Evolution. I will post the problems that I have had with it in the appropriate forum. However, I can't fix my mailbox now. When I change general.cfg and run omtidyu -B -u "Kenneth Lussier" -M i get:

Code: Select all

[root@postal klussier]# omtidyu -B -u "Kenneth Lussier" -M
Group 5 error calling ual_recvreply
Error reason 32
System error calling ual_sendcommand
Error reason 9
omtidyu : [OM.BP 2007] Protocol failure communicating with UAL server

the output of omscan isn't very helpful, either:

Code: Select all

[root@postal klussier]# omscan -Avx -U "Kenneth E. Lussier"
omscan running on 02.22.06 at 07:58:58.
Host computer : postal.
Report mode requested.

Last omscan tool run on 02.22.06 at 07:51:17; duration 1 minute(s).
Previous server cycle run on 02.16.06 at 18:07:14; duration 7 minute(s).
Current server cycle not started; service reset or delayed.


Extra check option requested.
Active scan option requested.   

Scanning file/dir links .... done.
                               
CAUTION: Scanning of message store has started.
         Mounted file/dir links must be maintained during the scan.
         VxFS file system must not be reorganized - see omscan(1M).
                               
Checking/Scanning user trays ....Kenneth E. Lussier / postal, aylus/CN=Kenneth Lussier
 done.                         
                               
Disk usage ....                 
                               
USER NAME                          IN    OUT    PDG   FCAB   DLST     WB   TOTAL (KB)
                               
Kenneth E. Lussier /postal,ayl  21157      1  23492  28713      1     47  73411


omsolve doesn't give me any information on the OM.BP error. Any ideas?

Thanks,
Kenny

[ScalixSupport]

Please also post the contents of /var/opt/scalix/logs/fatal.

Cheers

Dave

[kluss0]

Hi Dave,

Here is the fatal log output:

Code: Select all

SERIOUS ERROR           IMAP Server Da(IMAP Server Pr) Wed Feb 22 10:35:36 2006
[OM 10270] Process about to terminate due to error.
Signal (Segmentation Violation) trapped by process 10375
Procedure trace follows:
  -> imapMimeGetStructure
  -> imapMimeOpenTempFile
  <- imapMimeOpenTempFile
  -> ct_FindDITEntryByRef
  -> ct_DecodeDirectRef
  <- ct_DecodeDirectRef
  -> ct_FindDITEntry
  <- ct_FindDITEntry
  <- ct_FindDITEntryByRef
  -> msl_readContainerOpenedParent
  -> ct_ReadContentRec 3001 0
  <- ct_ReadContentRec 3001 0
  -> msl_reload
  -> ct_ReadExtension 3001 len=2504 rec=3164
  <- ct_ReadExtension
  <- ct_ReadExtension 3001 len=2504 rec=3164
User Name: Kenneth E. Lussier / postal, aylus/CN=Kenneth Lussier
Pid of logging process: 10375
  Last Msg Id: 1.1zpba.d4a.rs.0.4e
  Last Msg DirectRef: 000460ad6280b579


SERIOUS ERROR           IMAP Server Da(IMAP Server Pr) Wed Feb 22 10:35:36 2006
[OM 10272] BACKTRACE:
/opt/scalix/lib/libom_er.so(er_add_backtrace+0xc6)[0xd39f06]
/opt/scalix/lib/libom_er.so[0xd3a1d6]
/opt/scalix/lib/libom_er.so(er_DumpProcAndExit+0x1f)[0xd3a37f]
/lib/tls/libpthread.so.0[0xa838b8]
/opt/scalix/lib/libom_msl.so[0x19c66b]
/opt/scalix/lib/libom_msl.so[0x19c66b]
/opt/scalix/lib/libom_msl.so[0x19c778]
/opt/scalix/lib/libom_msl.so(msl_reload+0xa80)[0x19d200]
/opt/scalix/lib/libom_msl.so(msl_readContainerOpenedParent+0x857)[0x193b27]
in.imap41d[0x80620a5]
in.imap41d[0x805a9bb]
in.imap41d[0x805ef29]
in.imap41d[0x80602cd]
in.imap41d[0x8060b86]
/lib/tls/libc.so.6(__libc_start_main+0xe3)[0x285e33]
in.imap41d[0x804d981]
L|
User Name: Kenneth E. Lussier / postal, aylus/CN=Kenneth Lussier
Pid of logging process: 10375
  Last Msg Id: 1.1zpba.d4a.rs.0.4e
  Last Msg DirectRef: 000460ad6280b579


Thanks,
Kenny

[jasonesman]

Howdy,
We are also having this problem, we have narrowed it down to a multi part mime message. There is three parts and the third part is just an jpg. Once this message was removed this error didn't show anymore in the fatal log. This was also after trying to run omscan and omtidyallu -M. Here is the mime parts of the message

MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="----------=_43F3C7D6.4B5B803F"

This is a multi-part message in MIME format.

------------=_43F3C7D6.4B5B803F
Content-Type: text/plain
Content-Disposition: inline
Content-Transfer-Encoding: 8bit

------------=_43F3C7D6.4B5B803F
Content-Type: message/rfc822; x-spam-type=original
Content-Description: original message before SpamAssassin
Content-Disposition: attachment
Content-Transfer-Encoding: 8bit

MIME-Version: 1.0
Content-Type: multipart/related;
boundary="----=_NextPart_000_0051_01C631DB.DA3C46A0";

This is a multi-part message in MIME format.

------=_NextPart_000_0051_01C631DB.DA3C46A0
Content-Type: multipart/alternative;
boundary="----=_NextPart_001_0052_01C631DB.DA3C46A0"


------=_NextPart_001_0052_01C631DB.DA3C46A0
Content-Type: text/plain;
charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

------=_NextPart_001_0052_01C631DB.DA3C46A0
Content-Type: text/html;
charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

------=_NextPart_001_0052_01C631DB.DA3C46A0--

------=_NextPart_000_0051_01C631DB.DA3C46A0
Content-Type: image/jpeg;
name="38.jpg"
Content-Transfer-Encoding: base64
Content-ID: <005001c631a1$2ddafda0$6400a8c0@VAIO>

------=_NextPart_000_0051_01C631DB.DA3C46A0--

------------=_43F3C7D6.4B5B803F--

[kluss0]

I finally got the Scalix connector for Evolution to compile and install. Unfortunately, it apparently still uses IMAP because when I use a Scalix profile in Evolution, I get the error:

Code: Select all

Error while Opening folder scalix://klussier@postal.aylus.com/INBOX.
Server postal.aylus.com unexpectedly disconnected: No such file or directory

And I get all of the same IMAP errors on the server side.

[kluss0]

Based on the information in the above post, I used Outlook with the Scalix connector to copy all of the mail out of my inbox to a different folder. That seems to have fixed the problem. I can connect without any errors with SWA, any IMAP client, and Evolution using the Scalix connector. I can also open the folder that I copied all of my mail into without any errors, so it must have repaired the broken message.

FYI,
Kenny

[XJnash]

SERIOUS ERROR IMAP Server Da(IMAP Server Pr) Thu Nov 2 08:56:13 2006
[OM 10270] Process about to terminate due to error.
Signal (Segmentation Violation) trapped by process 3887
Procedure trace follows:
-> imapparse_address
<- imapparse_address
<- appendImapAddressList
-> appendImapAddressList
-> imapparse_address
<- imapparse_address
<- appendImapAddressList
-> appendImapAddressList
<- appendImapAddressList
-> appendImapAddressList
<- appendImapAddressList
<- appendEnvelope
-> appendBodyStructure
-> appendBodyStructure
<- appendBodyStructure
-> appendBodyStructure
User Name: Paul Rushing / post, highlandclinic/CN=Rushing, Paul
Pid of logging process: 3887


SERIOUS ERROR IMAP Server Da(IMAP Server Pr) Thu Nov 2 08:56:13 2006
[OM 10272] BACKTRACE:
/opt/scalix/lib/libom_er.so(er_add_backtrace+0xc6)[0x675f06]
/opt/scalix/lib/libom_er.so[0x6761d6]
/opt/scalix/lib/libom_er.so(er_DumpProcAndExit+0x1f)[0x67637f]
/lib/tls/libpthread.so.0[0x149898]
in.imap41d[0x805822f]
in.imap41d[0x80599d6]
in.imap41d[0x805aac5]
in.imap41d[0x805efa9]
in.imap41d[0x806034d]
in.imap41d[0x8060c06]
/lib/tls/libc.so.6(__libc_start_main+0xd3)[0x2afde3]
in.imap41d[0x804d9c1]
User Name: Paul Rushing / post, highlandclinic/CN=Rushing, Paul
Pid of logging process: 3887

I have followed the steps above and I'm still getting this error.

[kluss0]

Have you run an omscan against the mailstore? You may want to try:

Code: Select all

omscan -Avx -U Rushing

to see if any errors are reported. Can you post the output of omscan and omtidyu?


Kenny

[dkelly]

SERIOUS ERROR IMAP Server Da(IMAP Server Pr) Thu Nov 2 08:56:13 2006
[OM 10270] Process about to terminate due to error.
Signal (Segmentation Violation) trapped by process 3887
Procedure trace follows:
-> imapparse_address
<- imapparse_address
<- appendImapAddressList
-> appendImapAddressList
-> imapparse_address
<- imapparse_address
.
.

I have followed the steps above and I'm still getting this error.

You don't give any details on what the client is doing at the time. From the look of it, you are copying a message to a folder. Is this the case ?

Cheers

Dave