[asterisk-bugs] [JIRA] (ASTERISK-23990) [patch] app_voicemail crash when using IMAP greetings and forwarding messages

[Matt Jordan (JIRA)]

    [ https://issues.asterisk.org/jira/browse/ASTERISK-23990?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=220452#comment-220452 ] 

Matt Jordan commented on ASTERISK-23990:
----------------------------------------

Thanks for the backtrace.

If you'd like to post this up for code review for a more formal peer review, feel free to do so. Instructions for doing so are on the wiki here: https://wiki.asterisk.org/wiki/display/AST/Code+Review

> [patch] app_voicemail crash when using IMAP greetings and forwarding messages
> -----------------------------------------------------------------------------
>
>                 Key: ASTERISK-23990
>                 URL: https://issues.asterisk.org/jira/browse/ASTERISK-23990
>             Project: Asterisk
>          Issue Type: Bug
>      Security Level: None
>          Components: Applications/app_voicemail/IMAP
>    Affects Versions: 11.8.1
>            Reporter: Ben Smithurst
>            Assignee: Ben Smithurst
>         Attachments: app_voicemail_crash_fix.diff, bt.txt
>
>
> Asterisk currently crashes when trying to forward a message to another user - I think the combination of IMAP storage + IMAP greetings is the problem.
> What seems to happen is that Asterisk changes to the Greetings folder in the *source* mailbox instead of the destination.  It then doesn't change back, tries to retrieve message X (the one being forwarded) from Greetings, which doesn't exist, and libc-client calls abort().
> The reason appears to be due to the per-thread 'interactive' vmstate storage, Asterisk seems to assume that the per-thread requests will always be for the same mailbox, but when forwarding, that's not true.  So the fix is to check that the imapuser/mailbox/context (as appropriate) matches, when obtaining the interactive per-thread state.
> We can easily reproduce this problem, it appears to happen every time during our tests.



--
This message was sent by Atlassian JIRA
(v6.2#6252)