Open bugs to consider for 2.5

http://www.squid-cache.org/bugs/show_bug.cgi?id=14
attempts to establish connection look like syn flood attack
[this is due to our aggressive reopening of the connection on failure]

http://www.squid-cache.org/bugs/show_bug.cgi?id=81
basic scheme authentication helpers dont' allow the space character
[we need to decide if we should fix this and how, or just accept the
fact]

http://www.squid-cache.org/bugs/show_bug.cgi?id=94
assertion failed: http.c:869: "-1 == cfd || FD_SOCKET ==
fd_table[cfd].type"

http://www.squid-cache.org/bugs/show_bug.cgi?id=153
assertion failed: store_client.c when cache is full

http://www.squid-cache.org/bugs/show_bug.cgi?id=164
delay pools don't work with connection orientated auth
[investigate the correctness of the patch]

http://www.squid-cache.org/bugs/show_bug.cgi?id=207
Large responses may overflow some 32-bit counters on some systems
[patched in HEAD. Needs to be verified and then backported to 2.5]

http://www.squid-cache.org/bugs/show_bug.cgi?id=219
If there is active connections on reconfigure then these is no longer
delayed by their delaypools.

http://www.squid-cache.org/bugs/show_bug.cgi?id=250
Should use connected UDP sockets for DNS lookups

http://www.squid-cache.org/bugs/show_bug.cgi?id=263
Restart helpers if they die

http://www.squid-cache.org/bugs/show_bug.cgi?id=268
assertion failed: forward.c:84: "fwdState->err", shortly after a
parent went down.

http://www.squid-cache.org/bugs/show_bug.cgi?id=271
comm_quick_poll_required never called with diskd
[investigate and apply patch]

http://www.squid-cache.org/bugs/show_bug.cgi?id=272
Minor mismatch in Range forwarding

http://www.squid-cache.org/bugs/show_bug.cgi?id=301
To work around buggy servers we need to reopen the FTP data
connection on errors.

http://www.squid-cache.org/bugs/show_bug.cgi?id=327
Problems with IE browsers and persistent connections

Regards
Henrik
Received on Sat Apr 13 2002 - 09:37:23 MDT