| Commit message (Collapse) | Author | Age | Files | Lines |
| | |
|
| | |
|
| |\
| |
| | |
[1.2] Fix pending invites not being removed on channel deletion
|
| | |
| |
| |
| |
| |
| | |
TS lowered
Store pointers to channels instead of their names in User::invites
|
| | |
| |
| |
| |
| |
| | |
(retain old behavior)
Drop single targetted ENCAPs if target is reachable via the server who sent the command to us
|
| | |
| |
| |
| | |
source
|
| | |
| |
| |
| | |
with the protocol interface into the utility class
|
| | |
| |
| |
| | |
SpanningTreeProtocolInterface::SendUserPrivmsg/Notice, use utility functions like the rest of the protocol interface
|
| | |
| |
| |
| | |
printf'ing parameters
|
| | | |
|
| | |
| |
| |
| | |
incoming SVSJOIN/SVSPART but send it to the destination only, and don't pass it anywhere if they were talking to us
|
| | |
| |
| |
| |
| |
| |
| |
| | |
Most importantly:
- Match the server name not the SID for determining if an incoming ENCAP should be processed by us
- Handle the ? wildcard properly in the server name parameter of an ENCAP
- Send single target ENCAPs to the destination only instead of broadcasting it
- Make ProtocolInterface::SendMetaData accept TYPE_OTHER as target_type (on par with ProtoSendMetaData) instead of throwing an exception
|
| | |
| |
| |
| | |
client instead of sending PUSH commands manually
|
| |/ |
|
| | |
|
| |
|
|
|
|
| |
bans not the length of the parameters it queued for sending.
This has caused nearly every ban to be sent in a seperate FMODE.
|
| |
|
|
| |
IsBurstable() and not obeying loopCall, thus broadcasting a redundant DELLINE for every DELLINE received.
|
| | |
|
| |
|
|
| |
git-svn-id: http://svn.inspircd.org/repository/branches/1_2_stable@12647 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
| |
git-svn-id: http://svn.inspircd.org/repository/branches/1_2_stable@12511 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
| |
100% CPU in other socketengines
git-svn-id: http://svn.inspircd.org/repository/branches/1_2_stable@12473 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
| |
cause a double-free crash
git-svn-id: http://svn.inspircd.org/repository/branches/1_2_stable@12461 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
| |
servers
git-svn-id: http://svn.inspircd.org/repository/branches/1_2_stable@12458 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
| |
git-svn-id: http://svn.inspircd.org/repository/branches/1_2_stable@12393 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
| |
git-svn-id: http://svn.inspircd.org/repository/branches/1_2_stable@12391 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
| |
UID reuse on fast netsplits)
git-svn-id: http://svn.inspircd.org/repository/branches/1_2_stable@12327 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
| |
UID. Fixes bug #932
git-svn-id: http://svn.inspircd.org/repository/branches/1_2_stable@12319 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
| |
without SQUIT. Should fix the crash in bug #943
git-svn-id: http://svn.inspircd.org/repository/branches/1_2_stable@12316 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
|
|
| |
bug #928.
This will result in a mild desync if someone has (ab)used SVSJOIN when linking a server running this to their network, but since those channels are screwed anyway..
git-svn-id: http://svn.inspircd.org/repository/branches/1_2_stable@12234 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
|
|
|
| |
count if the user cant see these servers. and
calculate the user average over this number of servers too. -- Reported by Syloq, thanks!
git-svn-id: http://svn.inspircd.org/repository/branches/1_2_stable@11728 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
|
|
| |
collision on spanningtree reload") caused by access-after-free.
Thanks to transnet for the report.
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11533 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
| |
numbers (other server should fail the link if it becomes link-incompatable)
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11531 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
NEEDS COMPILE TESTING ON: gcc3 (if possible), gcc4, vs2008.
- Check for vs2010 in hash_map.h
- use unordered_map if it exists
- change all map creations to:
#if defined(WINDOWS) && !defined(HASHMAP_DEPRECATED)
// old windows crap
#else
#if HASHMAP_DEPRECATED
// tr1/gcc crap
#endif
#endif
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11524 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
| |
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11517 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
| |
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11438 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
|
| |
DNS requests (CNAME, A). It looks like this was to be related to bitmask
dnsbl support, but it would never have worked anyway (resultnum was always 0..).
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11419 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
| |
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11413 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
| |
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11404 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
| |
explosion of rehashes on a /REHASH *
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11388 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
| |
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11386 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
| |
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11385 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
| |
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11380 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
|
| |
This makes it possible to determine which server initiated a mode change,
instead of having only a flag saying that some server did.
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11379 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
| |
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11378 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
| |
fake user
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11377 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
| |
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11376 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
|
|
| |
This fixes some very subtle and hard-to-trace bugs that are triggered when a
file descriptor and memory address of an EventHandler* are reused after being
deallocated. Impossible to trigger in valgrind; has been seen in live networks.
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11369 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
| |
there should be an option to disable lag pings
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11331 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
| |
value; <1ms is misleading.
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11330 e03df62e-2008-0410-955e-edbf42e46eb7
|
| |
|
|
|
|
| |
build methods since the few of us doing win dev can't keep up maintaining the others.
git-svn-id: http://svn.inspircd.org/repository/trunk/inspircd@11327 e03df62e-2008-0410-955e-edbf42e46eb7
|