summaryrefslogtreecommitdiffstats
path: root/sbin/hastd/hastd.c
Commit message (Expand)AuthorAgeFilesLines
* Check remote protocol version only for the first connection (when ittrociny2013-12-101-6/+6
* - Add support for 'memsync' mode. This is the fastest replication mode that'spjd2013-02-171-2/+19
* If hastd is invoked with "-P pidfile" option always create pidfiletrociny2012-03-291-5/+7
* Free memory that won't be used in child.pjd2012-01-221-3/+7
* - Fix a bug where pidfile was removed in SIGHUP when it hasn't changed inpjd2012-01-101-4/+8
* For functions that return -1 on failure check exactly for -1 and not forpjd2012-01-101-20/+20
* Don't touch pidfiles when running in foreground. Before that change wepjd2012-01-101-11/+19
* Spelling fixes for sbin/uqs2012-01-071-1/+1
* Allow to specify pidfile in HAST configuration file.pjd2011-10-171-13/+61
* When path to the configuration file is relative, obtain full path,pjd2011-10-171-0/+16
* After every activemap change flush disk's write cache, so that writepjd2011-09-281-5/+16
* In preparation for IPv6 support allow to specify multiple addresses topjd2011-05-191-36/+97
* Rename HASTCTL_ defines, which are used for conversion between maintrociny2011-04-261-1/+1
* Correct comment.pjd2011-04-201-1/+1
* When we become primary, we connect to the remote and expect it to be inpjd2011-04-201-0/+7
* If we act in different role than requested by the remote node, log itpjd2011-04-201-1/+1
* Don't create socketpair for connection forwarding between parent and secondary.pjd2011-03-231-8/+28
* White space cleanups.pjd2011-03-221-6/+6
* Before handling any events on descriptors check signals so we can updatepjd2011-03-211-28/+43
* In hast.conf we define the other node's address in 'remote' variable.pjd2011-03-211-1/+9
* When creating connection on behalf of primary worker, set pjdlog prefixpjd2011-03-211-0/+4
* If there is any traffic on one of out descriptors, we were not checking forpjd2011-03-211-2/+14
* Allow to compress on-the-wire data using two algorithms:pjd2011-03-061-0/+6
* Allow to checksum on-the-wire data using either CRC32 or SHA256.pjd2011-03-061-1/+7
* Now that we break the loop on fstat(2) failure we no longer need to satisfypjd2011-02-061-2/+0
* Add (void) cast before snprintf(3)s for which we are not interested in returnpjd2011-02-061-13/+13
* Treat fstat(2) failure (different than EBADF) as fatal error.pjd2011-02-061-0/+1
* Open syslog when logging sysconf(3) failure.pjd2011-02-061-0/+4
* Close more descriptors that can be open if the worker process for the givenpjd2011-02-061-0/+6
* Setup another socketpair between parent and child, so that primary sandboxedpjd2011-02-031-0/+64
* - Use pjdlog for assertions and aborts as this will log assert/abort messagepjd2011-01-311-15/+14
* Add function to assert that the only descriptors we have open are the onespjd2011-01-281-0/+141
* Add function to close all unneeded descriptors after fork(2).pjd2011-01-281-1/+27
* Close the control socket before exiting, so it will be unlinked.pjd2011-01-271-0/+1
* Add functions to initialize/finalize pjdlog. This allows to open/close logpjd2011-01-271-0/+1
* Don't open configuration file from worker process. Handle SIGHUP in thepjd2011-01-241-7/+41
* - On primary worker reload, update hr_exec field.pjd2011-01-221-2/+5
* Install default signal handlers before masking signals we want to handle.pjd2011-01-121-0/+8
* Log the fact of launching and include protocol version number.pjd2010-12-161-0/+3
* Load geom_gate.ko module after parsing arguments.pjd2010-10-241-2/+2
* Log correct connection when canceling half-open connection.pjd2010-10-171-1/+2
* Decrease report interval to 5 seconds, as this also means we will check forpjd2010-10-041-1/+1
* hook_check() is now only used to report about long-running hooks, so thepjd2010-10-041-1/+1
* We can't mask ignored signal, so install dummy signal hander for SIGCHLD beforepjd2010-10-041-0/+11
* Switch to sigprocmask(2) API also in the main process and secondary process.pjd2010-09-221-45/+39
* Assert that descriptor numbers are sane.pjd2010-09-221-0/+4
* Fix descriptor leaks: when child exits, we have to close control and eventpjd2010-09-221-8/+2
* Because it is very hard to make fork(2) from threaded process safe (we arepjd2010-08-301-10/+34
* We only want to know if descriptors are ready for reading.pjd2010-08-301-7/+4
* Allow to run hooks from the main hastd process.pjd2010-08-291-3/+17
OpenPOWER on IntegriCloud