- 17 Jan, 2012 1 commit
-
-
JINMEI Tatuya authored
the detailed version of comments were intended to be for auth, but placed in the wrong place (resolver/main).
-
- 11 Jan, 2012 5 commits
-
-
JINMEI Tatuya authored
talked to the boss (via msgq), not the socket creator directly). also consolidated the mostly identical comments for auth and resolver for DRY.
-
JINMEI Tatuya authored
-
JINMEI Tatuya authored
to clarify that it's also necessary for listen_on.
-
JINMEI Tatuya authored
until the initial configuration is completed. otherwise it confuses communication with the socket creator and makes listen_on fail.
-
JINMEI Tatuya authored
-
- 10 Jan, 2012 1 commit
-
-
Michal 'vorner' Vaner authored
-
- 09 Jan, 2012 2 commits
-
-
Michal 'vorner' Vaner authored
-
Michal 'vorner' Vaner authored
-
- 07 Jan, 2012 3 commits
-
-
-
JINMEI Tatuya authored
-
JINMEI Tatuya authored
more strict on this point.
-
- 06 Jan, 2012 8 commits
-
-
JINMEI Tatuya authored
it's 0 or non 0). same change for IPV6_V6ONLY, too. also change ASSERT_EQs to EXPECTs: failure of them doesn't seem to cause catastrophy in the subsequent tests. made also some style guideline fixes, too.
-
JINMEI Tatuya authored
close the function if send_fd fails.
-
JINMEI Tatuya authored
value identical to the parameter name itself.
-
Michal 'vorner' Vaner authored
-
JINMEI Tatuya authored
If it is run from the source code, all programs should be run from there.
-
Michal 'vorner' Vaner authored
Added the v6 only flag for IPv6 sockets. Both reuse address and this flags get tested.
-
Michal 'vorner' Vaner authored
-
Michal 'vorner' Vaner authored
-
- 05 Jan, 2012 8 commits
-
-
Michal 'vorner' Vaner authored
* Close the socket after sending from the creator, so it doesn't leak (but it doesn't seem to help :-(). * Correct range for the token numbers, so they are harder to guess * Set the socket as reuse address
-
Michal 'vorner' Vaner authored
-
Michal 'vorner' Vaner authored
-
Michal 'vorner' Vaner authored
The socket and an address. We don't care about the address, but we still need to unpack the socket out.
-
Michal 'vorner' Vaner authored
The ModuleCCSession called the update of config which already needed the SocketRequestor. But the requestor doesn't need the ModuleCCSession, cc::Session is enough, so changing it and moving up in the initialization process a little.
-
Michal 'vorner' Vaner authored
-
Michal 'vorner' Vaner authored
Conflicts: src/lib/server_common/socket_request.cc src/lib/util/io/fd.cc
-
Michal 'vorner' Vaner authored
-
- 04 Jan, 2012 12 commits
-
-
JINMEI Tatuya authored
temporary variable.
-
JINMEI Tatuya authored
other comment on this point.
-
JINMEI Tatuya authored
-
JINMEI Tatuya authored
-
Jeremy C. Reed authored
slight rewording made a preface
-
Michal 'vorner' Vaner authored
It was missing an EXTRA_DIST entry to supply the test spec data
-
Michal 'vorner' Vaner authored
-
Jeremy C. Reed authored
moved some content changed some formatting tags mention more platforms add some anchor link IDs add some URls for dependencies slight rewording point to wiki for more tips
-
Michal 'vorner' Vaner authored
-
Michal 'vorner' Vaner authored
-
JINMEI Tatuya authored
By having the TestSocketRequestor as member.
-
Michal 'vorner' Vaner authored
The service is now (nonstatic) member and the static variable is only pointer to the active one, so the signal handler can access it.
-