129226 - Memory leak while running tstclnt on BoundsChecker - allocated by strdup in ssl3con.c

Status: RESOLVED DUPLICATE of bug 133584

(NSS :: Libraries, defect, P2)

Description

[Bishakha Banerjee]

Call stack as follows:

Memory leak
10 bytes allocated by strdup in ssl3con.c (4770), HANDLE: 0x01A50F00

Location of Error
ssl3_NewSessionID
ssl3con.c
4770
ssl3_HandleServerHello
ssl3con.c
4145
ssl3_HandleHandshakeMessage
ssl3con.c
7139
ssl3_HandleHandshake
ssl3con.c
7258
ssl3_HandleRecord
ssl3con.c
7523
ssl3_GatherCompleteHandshake
ssl3gthr.c
203
ssl_GatherRecord1stHandshake
sslcon.c
1256
ssl_Do1stHandshake
sslsecur.c
155
ssl_SecureSend
sslsecur.c
1036
ssl_SecureWrite
sslsecur.c
1070
ssl_Write
sslsock.c
1260
PR_Write
priometh.c
141
main
tstclnt.c
670
_mainCRTStartup
TSTCLNT.EXE!00007EFD
(unknown)

Comment 1

[Wan-Teh Chang]

Nelson, do you know why we are leaking the 'urlSrvName' field of
a sslSessionID structure on the client side?  The only function
I can find that destroys sslSessionID is ssl_FreeSID, which
frees sid->urlSrvName (in ssl_DestroySID).  So I have no idea why
we are leaking this field.

Comment 2

[Nelson Bolyard (seldom reads bugmail)]

Unless the program has been modified to flush the SSL client SID cache,
I suspect that all the SIDs in the client SID cache will appear to have
been "leaked" at the end of the program.  If the program only does one
https transaction, there will only be one SID that is leaked.  

If you want to clear the client SID cache, call SSL_ClearSessionCache(void)

Comment 3

[Wan-Teh Chang]

Changed the QA contact to Bishakha.

Comment 4

[Wan-Teh Chang]

Set target milestone to NSS 3.5.

Comment 5

[Wan-Teh Chang]

Moved to target milestone 3.8 because the original
NSS 3.7 release has been renamed 3.8.

Comment 6

[Nelson Bolyard (seldom reads bugmail)]

Remove target milestone of 3.8, since these bugs didn't get into that release.

Comment 7

[Julien Pierre]

This was fixed as part of bug 133584 . tstclnt was made to call SSL_ClearSessionCache() .