124846 - ASSERTION: object should be closed: 'obj->closed_p'

Status: NEW

(MailNews Core :: MIME, defect)

Description

[timeless]

trying to read ssaux's nss3.4 npm.reviewers requests (1/30/02 22:40)
nsNativeComponentLoader: GetFactory(/home/timeless/mozilla/obj-gtk-i386-unknown-freebsd4.4/dist/bin/components/libpipnss.so) Load
 FAILED with error: Cannot open "../../../../dist/lib/libsmime3.so"
**************************************************
###!!! ASSERTION: object should be closed: 'obj->closed_p', file /home/timeless/mozilla/mailnews/mime/src/mimeobj.cpp, line 300
###!!! Break: at file /home/timeless/mozilla/mailnews/mime/src/mimeobj.cpp, line 300

#1  0x282db763 in nsDebug::Break (aFile=0x2a78b840 "/home/timeless/mozilla/mailnews/mime/src/mimeobj.cpp",
    aLine=300) at /home/timeless/mozilla/xpcom/base/nsDebug.cpp:322
#2  0x282db645 in nsDebug::Assertion (aStr=0x2a78baf0 "object should be closed",
    aExpr=0x2a78bae2 "obj->closed_p",
    aFile=0x2a78b840 "/home/timeless/mozilla/mailnews/mime/src/mimeobj.cpp", aLine=300)
    at /home/timeless/mozilla/xpcom/base/nsDebug.cpp:291
#3  0x2a7445da in MimeObject_parse_end (obj=0x9434280, abort_p=0)
    at /home/timeless/mozilla/mailnews/mime/src/mimeobj.cpp:300
#4  0x2a731ca8 in MimeContainer_parse_end (object=0x9434280, abort_p=0)
    at /home/timeless/mozilla/mailnews/mime/src/mimecont.cpp:160
#5  0x2a743f49 in MimeObject_finalize (obj=0x9434280)
    at /home/timeless/mozilla/mailnews/mime/src/mimeobj.cpp:148
#6  0x2a731ba1 in MimeContainer_finalize (object=0x9434280)
    at /home/timeless/mozilla/mailnews/mime/src/mimecont.cpp:116
#7  0x2a74285c in MimeMultipart_finalize (object=0x9434280)
    at /home/timeless/mozilla/mailnews/mime/src/mimemult.cpp:136
#8  0x2a7411b3 in MimeMultipartSigned_finalize (obj=0x9434280)
    at /home/timeless/mozilla/mailnews/mime/src/mimemsig.cpp:174
#9  0x2a738c3d in mime_free (object=0x9434280) at /home/timeless/mozilla/mailnews/mime/src/mimei.cpp:280
#10 0x2a731b40 in MimeContainer_finalize (object=0x9432080)
    at /home/timeless/mozilla/mailnews/mime/src/mimecont.cpp:110
#11 0x2a73f691 in MimeMessage_finalize (object=0x9432080)
    at /home/timeless/mozilla/mailnews/mime/src/mimemsg.cpp:123
#12 0x2a738c3d in mime_free (object=0x9432080) at /home/timeless/mozilla/mailnews/mime/src/mimei.cpp:280
#13 0x2a74f312 in mime_display_stream_complete (stream=0x90fc840)
    at /home/timeless/mozilla/mailnews/mime/src/mimemoz2.cpp:909
#14 0x2a75d123 in nsStreamConverter::OnStopRequest (this=0x9427a80, request=0x8e6ba10, ctxt=0x943b100,
    status=0) at /home/timeless/mozilla/mailnews/mime/src/nsStreamConverter.cpp:1010
#15 0x28e613db in nsDocumentOpenInfo::OnStopRequest (this=0x90e8780, request=0x8e6ba10, aCtxt=0x943b100,
    aStatus=0) at /home/timeless/mozilla/uriloader/base/nsURILoader.cpp:253
#16 0x28bf0ac4 in nsStreamListenerTee::OnStopRequest (this=0x90c9d20, request=0x8e6ba10, context=0x943b100,
    status=0) at /home/timeless/mozilla/netwerk/base/src/nsStreamListenerTee.cpp:24
#17 0x2aa210c7 in nsNNTPProtocol::CleanupAfterRunningUrl (this=0x8e6ba00)
    at /home/timeless/mozilla/mailnews/news/src/nsNNTPProtocol.cpp:5424
#18 0x2aa20c4c in nsNNTPProtocol::ProcessProtocolState (this=0x8e6ba00, url=0x8924004, inputStream=0x8d52c90,
    sourceOffset=42482, length=6281) at /home/timeless/mozilla/mailnews/news/src/nsNNTPProtocol.cpp:5352
#19 0x2a6cb5c9 in nsMsgProtocol::OnDataAvailable (this=0x8e6ba0c, request=0x8e56200, ctxt=0x8924000,
    inStr=0x8d52c90, sourceOffset=42482, count=6281)
    at /home/timeless/mozilla/mailnews/base/util/nsMsgProtocol.cpp:264
#20 0x28bef85d in nsOnDataAvailableEvent::HandleEvent (this=0x94966c0)
    at /home/timeless/mozilla/netwerk/base/src/nsStreamListenerProxy.cpp:193
#21 0x28bda408 in nsARequestObserverEvent::HandlePLEvent (plev=0x94966c0)
    at /home/timeless/mozilla/netwerk/base/src/nsRequestObserverProxy.cpp:115

Yes I know the smime libs failed, that's not my fault, and this is about fault tolerance.

Comment 1

[David :Bienvenu]

looks like a mime problem or an smime problem...-> JF, cc mscott.

Comment 2

[Stephane Saux]

Isn't this a dupe of bug 124211 ?
cc kai.
 

Comment 3

[Kai Engert [:KaiE:]]

I agree the cause of this problem is bug 124211. But Timeless said he complains
about fault tolerance.

If we want to detect whether a component is missing or not, should we try to
call a dummy method in each necessary component at startup?

Currently, we make the assumption:
  if libpipnss.so is available => assume crypto is working

Now that NSS is no longer an internal part of libpipnss.so, but resides in
external shared libraries, this assumption is no longer true.

We already have a mechanism inside PSM, being able to check at load time,
whether everything looks good or not. If not, we refrain from offering any
crypto services to the Mozilla application. We could extend that checking
mechanism to call check methods from each NSS library.

Comment 4

[Wan-Teh Chang]

Kai wrote:
> We already have a mechanism inside PSM, being able to
> check at load time, whether everything looks good or not.
> If not, we refrain from offering any crypto services to
> the Mozilla application. We could extend that checking
> mechanism to call check methods from each NSS library.

It is not necessary to call check methods from each NSS
library.  If the NSS libraries are not availble, libpipenss.so
can't be loaded.  (See the error message from
nsNativeComponentLoader in the bug description.)  Conversely,
if you have successfully loaded libpipenss.so, you know the
NSS libraries are available.