IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
network connectivity outside of the cluster to still be able to
participate in a natgw group.
These nodes can not become natgw master since they lack external network
connectivity.
These nodes are configured just the same way as for any other node with
NATGW, with the following two exceptions :
* we do NOT set CTDB_NATGW_PUBLIC_IFACE at all on these nodes.
since these ndoes lack external network we should not check the interface
for link.
* we must set CTDB_NATGW_SLAVE_ONLY=yes to flag that this is a node that
can not become natgw master.
(This used to be ctdb commit ab7b00a37e55beffc074be95b55d8a5c7cb9eef2)
since that will usually be /etc/ctdb/state and storing this under /etc is just
wrong.
Add a new variable CTDB_VARDIR that defaults to /var/ctdb and store the data there instead.
(This used to be ctdb commit 516423c25afa9861d9988096efa8a4a2b12b31b1)
the clusterwide persistent data associated with the lock manager and
statd notifications.
Use persistent databases to store this data instead of a shared directory.
(This used to be ctdb commit fc0678d351187cfa4c71123f97c0f493aacd5d16)
This is called everytime a reallocation is performed.
While STARTRECOVERY/RECOVERED events are only called when
we do ipreallocation as part of a full database/cluster recovery,
this new event can be used to trigger on when we just do a light
failover due to a node becomming unhealthy.
I.e. situations where we do a failover but we do not perform a full
cluster recovery.
Use this to trigger for natgw so we select a new natgw master node
when failover happens and not just when cluster rebuilds happen.
(This used to be ctdb commit 7f4c591388adae20e98984001385cba26598ec67)
This adds a new function update_tickles() that tracks tickles for a
given port using the new ctdb addtickle/deltickle commands. This
function is used in events.d/60.nfs to handle NFS tickles.
events.d/61.nfstickle is removed. The
/proc/sys/net/ipv4/tcp_tw_recycle setup is also moved to
events.d/60.nfs.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit dca4c4ebf3c35f8db3ae208efb7a83abbf726ed6)
This database can be used, as an option, to store
the public address assignment instead of editing the /etc/ctdb/public-addresses file manually.
This configuration is stored in one record per key, with a key-name of
public-addresses:node#<pnn>
where <pnn> is the node number.
The content of this record is the same syntax as the /etc/ctdb/public-addresses file.
When ctdbd starts, if this key exist and contains data. It is extracted from the database and compared with the normal file /etc/ctdb/public-addresses.
If the content differs, the config database "wins" and is used to overwrite/update the /etc/ctdb/public-addresses file, after which ctdbd is restarted.
The main benefit with this option is that it can be used to update the public address configuration for nodes that are offline/unreachable by updating their configuration in the persistent database.
Once the offline node is available again, it will resync its databases with the rest of the cluster, find out that the config has changed, apply the changes and restart ctdbd automatically.
The command to store the public address configuration for a node into the persistent database is :
ctdb pstore config.tdb public-addresses:node#<pnn> <filename>
where <pnn> is the node# we wish to update the config for, and <filename> is a file containing the new content for that nodes public address configuration.
(This used to be ctdb commit 292d7435a360efd7f15a7a99f658a605e07c0a81)
sometimes (very rarely) fails to restart the service.
Add a function to restart NFSd on SLES and RHEL-like systems.
If we detect the system is unhealthy due to kNFSd not running,
try to restart the service again "service nfs restart" and
hope for the best.
CQ1019372
(This used to be ctdb commit 25c4ce7e919f13226219f036bcffd2be76b2f06c)
The existing code wasn't working as designed in the start event. It
should work here.
BZ: 62613
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit aeb70c7e7822854eb87873a5c7783e27e6e72318)
Currently we do a "sleep 1" after starting and before running
set_ctdb_variables to set the tunables. This is too arbitrary and
might fail if the system is heavily loaded. This, for example, could
result in some nodes running with DeterministicIPs and some without,
in which case a different IP allocation algorithm would run depending
on who is the recmaster!
This makes the start function wait until "ctdb ping" succeeds (with 10
second timeout) before trying to run set_ctdb_variables. If a timeout
occurs then the start function attempts to kill ctdbd before exiting
with a failure.
It also cleans up the status reporting code for Red Hat and SUSE so
that the final status code is reported. Currently there are cases
where a correct status is prematurely reported before a failure
occurs.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit cdcd05662a30b51caaeeab4ac44138cac2474e0a)
Currently the file for each IP address is reopened to append the
details of each source socket.
This optimisation puts all the logic into awk, including the matching
of output lines from netstat. The source sockets for each for each
destination IP are written into an array entry and then each array
entry is written to the corresponding file in a single operation.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 6549e9b01538998d51a5f72bfc569776d232b024)
For non bash shells $_s_script might end with '/*'.
We do the workarround this way, because it makes sense to check
that a script is executable, before trying to execute it.
metze
[ This actually applies to any shell -- Rusty Russell ]
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
(This used to be ctdb commit e665cfde03fc9ec2264e99512ed5470872a2fd04)
When doing a releaseip event, we do them in parallel for all the separate
IPs. This creates a problem for iptables, which isn't reentrant, giving
the strange message:
iptables encountered unknown error "18446744073709551615" while initializing table "filter"
The worst possible symptom of this is that releaseip won't remove the rule
which prevents us listening to clients during releaseip, and the node will be
healthy but non-responsive.
The simple workaround is to flock-wrap iptables. Better would be to rework
the code so we didn't need to use iptables in these paths.
CQ:S1018353
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
(This used to be ctdb commit 72d6914ee913272312d7b68f1be5ad05ad06587d)
same ip address as a normal public-address,
check for this in the natgw script and warn the user.
Also prevent ctdb from starting up since this configuration will not work.
BZ60933
(This used to be ctdb commit 480af69b63b9162c85d8e04461ca9e4a083c04a4)
If the driver is virtio_net then we assume that the link is up rather
than ignoring the check altogether.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 3044d07da2a58260fa06bf489890b279bcf3ec39)
Skip link test for this type of devices
Signed-off-by: Ralph Wuerthner <ralph.wuerthner@de.ibm.com>
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 2ea0a9f1a93781a0d036feb9fcc0d120b182922f)
As the basename of the script will be used for the readd script
from setup_iface_ip_readd_script, it's know easier to identify
what script is called by delete_ip_from_iface() while readding
ips to the interface.
metze
(This used to be ctdb commit 3ee225b0b6ed37c22478bd145ced56b1b9b86842)
This is needed because we need to resetup the routing table when
the delete_ip_from_iface() function readds the ip to the interface.
metze
(This used to be ctdb commit ea87185ec9977006ef72d5a68c875154e4c84099)
This combines the logic into a shell function which can be used by the
"takeip" and "updateip" hooks.
We check the return values of the "ip" commands now
instead of ignoring them.
We now create a setup_script.sh similar to the release_script.sh
which makes it easier to analyze problems.
metze
(This used to be ctdb commit 624e8878851b4957cc7c02e922ec86926d6927ee)
This also initializes the variables correctly for the
shutdown|removenatgw code path to delete_all.
metze
(This used to be ctdb commit 2c2cbed4fcbc868a990fa6b32fc96126ffc61bb5)
This adds a generic infrastructure to register scripts which will
be called when the delete_ip_from_iface() funtion needs to readd
secondary ips to an interface.
metze
(This used to be ctdb commit ac97d65f44e8dc8bf2ec8f68e4db3448521755a2)
to control whether or not to check if we are swapping, and produce
useful output into the logfile if we are.
For production systems with dedicated nas-heads we should never swap.
But for developer/test systems we often use smaller nondedicated systems where
we can no longer guarantee that we will not be using swap.
(This used to be ctdb commit db87849bf3380914a63a626412bec209dbea7d20)
We should never enter swap; if we do, show the memory state of the machine and the process list. This will help us diagnose what caused the condition before it's too late and the box starts OOM-killing processes.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
(This used to be ctdb commit 627a6d67a0e9e61f8713e62695b3518c51909230)
If "$1" was empty than loadconfig would load the ctdb config twice.
This stops that from happening.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 0406d406da70aaee7ad6aac236114905c5d03ed2)
Proper fix for 085d1bea78fabf754ef6dd6d323f74a1d361e45c's workaround.
$NFS_TICKLE_SHARED_DIRECTORY was being used before it is set via
loadconfig.
Ronnie actually spotted this one. :-)
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit ee8b2e298351d05197a2e1494f3331433644c1e6)
Also, change the order of the comparison so it is consistent with
others in the script.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 44696e15cdb23e7656d3bb0ead54f509495738a7)
This puts single quotes around everything and uses eval on the
command-lines that actually start ctdbd. The eval causes the single
quotes to be interpreted.
The "redhat" init style no longer uses the Red Hat daemon function.
It loses the quoting and re-splits on spaces. Instead we add an extra
line that uses the success/failure functions to keep things pretty.
Note that this means that we don't respect daemon's
$DAEMON_COREFILE_LIMIT variable but we do our own core file handling
with $CTDB_SUPPRESS_COREFILE anyway. daemon's core file handling was
probably overriding what we were doing anyway, so this can be regarded
as a bug fix.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 522fbb012524fe41a67dbe43589a282dda6bcbe2)
This is very useful for testing, I use such a script:
cat ~/bin/ethtool
#!/bin/sh
IFACE=$1
case "$IFACE" in
Neth2)
;;
Neth3)
;;
Neth4)
;;
Neth5)
;;
*)
exec /usr/sbin/ethtool $@
;;
esac
ip link set down $IFACE
exec /usr/sbin/ethtool $@
metze
(This used to be ctdb commit 3bab985cf615720eded4d47b4f9f37a9c28840aa)
With this option set to "yes", we don't become unhealthy
as long as at least one interface is still available.
metze
(This used to be ctdb commit d054eb33c6ae92560cddb40732e5dcf622591a3c)
With this script it's possible to generate routing tables
per public ip address.
metze
(This used to be ctdb commit ff5678fbec2daef461143acf00cef3f94d7655fc)
When two releaseip events run in parallel it's possible that the 2nd script
readds a secondary ip that was removed by the 1st script.
metze
(This used to be ctdb commit e02417b2a55c45ac2c125b1b3463c9c39e7bc07a)
This is needed because the "startup" event runs after the initial recovery,
but we need to do some actions before the initial recovery.
metze
(This used to be ctdb commit e953808449c102258abb6cba6f4abf486dda3b82)
configureable using --log-ringbuf-size=<num-entries>.
Add an entry in the sysconfig file to set this persistently.
(This used to be ctdb commit c79c2da69bc352f509e7fca4b9172a4b7f23c0f8)
This reverts commit 7c95e56ba871a4e0cb893a5cb5d821e7ff6e6dd6.
wbinfo --ping-dc is proving too unreliable.
(This used to be ctdb commit b70021856e76df1ba407c83cfc19bf332fbfc869)
This reverts commit 7b73834ba3ac197cc8a3020c111f9bb2c567e70b.
wbinfo --ping-dc is proving too unreliable.
(This used to be ctdb commit 178f429a7b6d1008d35e857b6ca1df6adb60d255)
The do_setsched was being tested for whether to mmap tdbs: let's make it
explicit. We can also happily move the kill-child eventscript hack under
this flag.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
(This used to be ctdb commit 2ee86cc1f311d7b7504c7b14d142b9c4f6f4b469)
All other scripts do 'loadconfig ctdb' before any other 'loadconfig foo'
call. I think we should do the same in statd-callout.
Otherwise it's very confusing, if you have configured some Options
in /etc/sysconfig/ctdb, but /etc/ctdb/statd-callout doesn't notice
them.
metze
(This used to be ctdb commit 10d95581fb90bfdf58ec32345c4e36c27acf4f37)
when checking link status for an interface, first
check if this interface is in fact a bond device
(by the precense of a /proc/net/bonding/IFACE file)
and use that file for checking status.
Othervise assume ib* is an infiniband interface which we donnt know how
to check, or otherwise it is an ethernet interface and ethtool should
hopefully work.
(This used to be ctdb commit 8cc6c5de3d7abb0b72eaa6e769e70963b02d84cb)
The functions file no longer causes a side-effect by doing a shift.
It also doesn't set a convenience variable for $1.
All eventscripts now explicitly use "$1" in their case statement, as
does the initscript. The absence of a shift means that the
takeip/releaseip events now explicitly reference $2-$4 rather than
$1-$3.
New function ctdb_standard_event_handler handles the status and
setstatus events, and exits for either of those events. It is called
via a default case in each eventscript, replacing an explicit status
case where applicable.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 3d55408cbbb3bb71670b80f3dad5639ea0be5b5b)
Apart from lots of cleanup work, this also fixes a bug where the share
checks didn't used to cope with directory names containing spaces.
The previous commit also loaded the config incorrectly.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 3c93336ab92c2e4829ff4dc360045bfa6df21d50)
This is the first stage of an experimental change to eventscripts.
Ronnie and I did a few hours of factorisation of 40.vsftpd and applied
many of the changes to 41.httpd. Other eventscripts were also
modified.
At this stage this is completely untested.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 364e70b763f0ccd7714d15723ad3ea4d7e2968a1)
use killtcp and kill both directions of the nfs connections.
we used to kill only one direction since the other direction was unkillble
but recent kernels allow us to kill both
(This used to be ctdb commit 8001ae580bcc28d45f6026b529d7ffc247cbba34)
This just sleeps for twice the value of EventScriptTimeout
in the monitor action. It is not run by default, but
can be activated by setting CTDB_RUN_TIMEOUT_MONITOR
in /etc/sysconfig/ctdb .
Michael
(This used to be ctdb commit 1a3ecdee85b82bb3234a92ae6bcdeb92238eb7ee)
It's much nicer for post-mortem debugging to have a body to examine.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
(This used to be ctdb commit 058e21d96c3c02759833fd5ddfe7b43e6a5f5740)
this to handle the case where all links do have a physical layer, but where all slaves have been disabled using ifdown
(This used to be ctdb commit bf50709630df000583f2b0ef0edc177c01d60eaf)
Leave the node as UNHEALTHY this stops clients from accessing the node until
the reclock file can be accessed again
(This used to be ctdb commit f5e9f3007c10a937158bc8cdfabf33c984cf9c50)
this allows us to configure and enable nfs at runtime without having to restart ctdbd
(This used to be ctdb commit f6e39d35713475defaa08a623e194f3f2f8f7d53)
There are 2 problems with this code:
* The loop in ctdb_check_directories_probe() breaks on filenames
containing whitespace.
The fix to protect them is to pass "$@" to this function and have it
operate on "$@".
Note that there's still a problem with whitespace in filenames in
the 50.samba eventscript. To fix this ctdb_check_directories_probe
should read the filenames from stdin. Another time...
* The check for '%' in filenames in ctdb_check_directories_probe()
ends up involving several forks. On a modern machine this can cost
a couple of minutes when checking a large number of directories.
The fix is to use a case statement.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit eb1fecaef9aa5cb85dff7d4f7af8a9878deabed8)
Each recovery that involves IP reassignments results in a restart of
vsftpd in the "recovered" event. Currently, we can have several
recoveries in quick succession and the "monitor" event following each
can fail because vsftpd isn't ready yet. This results in cumulative
failures, so the node is marked unhealthy, even though vsftpd has
never had a proper opportunity to become ready.
This resets the fail count after each recovery.
While we're here, also move the delete of the restart flag file into
the body of the conditional.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 318abeb4b913a8d846e7eaf4cf5c2a67b61ce974)
test -z really needs its argument to be quoted. Simplified a status
test.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit fe26da7780545b1ecc0a7da5bc1cf8beaeea94cc)
Change the monitor event in 40.vsftpd so it only fails if there are 2
successive failures connecting to port 21. This reduces the
likelihood of unhealthy nodes due to vsftpd being restarted for
reconfiguration due to node failover or system reconfiguration.
New eventscript functions ctdb_counter_init, ctdb_counter_incr,
ctdb_counter_limit. These are used to count arbitrary things in
eventscripts, depending on the eventscript name and a tag that is
passed, and determine if a specified limit has been hit. They're good
for counting failures!
These functions are used in 40.vsftpd and also in 01.reclock - the
latter used to do the counting without these functions.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit cfe63636a163730ae9ad3554b78519b3c07d8896)
Remove the explicit vacuum/repack commands from the 00.ctdb eventscript
and implement this in the ctdb daemon.
Combine vacuuming and repacking into one
cheap read traverse to enumerate all candidate records
and one write traverse that both repacks the database and also deletes the record locally where we are lmaster and where the records have already been deleted remotely.
this code also adds initial autotuning heuristics for the vacuum intervals and how many records to delete in each iteration.
minor stylish changes made by ronnie s
(This used to be ctdb commit 95a3ee551241aa164967991fe5efe078e1714bde)
if the reclock file has been set, then this script will test that the
reclock file can actually be accessed.
if the file does not exist, or if the attempts to stat the file hangs,
the node will be marked unhealthy after the third failed monitoring event
and after the tenth failure, ctdb itself will shutdown.
(This used to be ctdb commit 2cb04747887674def299e574fccb827c1c3194e7)
This allows for controlling start of ctdbd with or without the option "--syslog"
from the sysconfig/ctdb file.
Michael
(This used to be ctdb commit 7bf9fff9139a4270496bddb97f9433bab87824bf)
dont log errors is trying to delete a nonexisting state file
this eliminates some annoying log entries in the ctdb log
(This used to be ctdb commit 7a95257a5ec19f232f661bc7f797051bf08ab776)
This does not modify any behaviour of the daemon itself other than showing this flag as ON in the ctdeb getcapabilities output
(This used to be ctdb commit fb337c151bd16ad5ad0c99431224451979d8c651)
This event is called when a node is stopped and is used by eventscripts that need to do certain cleanup and removal of configuration or ip addresses or routing ...
Note that a STOPPED node is considered "inactive" and as such will not be running the "recovered" event when the rest of the cluster has recovered.
(This used to be ctdb commit 65e9309564611bf937ded3c74a79abff895d7c59)
also verify that we actually do have a natgw master available if this is configured and make the node unhealthy if not.
(This used to be ctdb commit 7f273ee769d671d8c8be87c9187302fb77e814f3)
* Move building of CTDB_OPTIONS to new function build_ctdb_options()
and have it use a helper function for readability.
* New functions check_persistent_databases() and set_ctdb_variables().
* Remove valgrind-specific stop code, since the general pkill should
kill ctdbd when running under valgrind.
* Remove some bash-isms (e.g. >& /dev/null) since the script is /bin/sh.
* Make indentation consistent.
* Minor clean-ups.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Conflicts:
config/ctdb.init
(This used to be ctdb commit bebb21f18e3026cb78a306104e92ee005d1077b2)
The valgrind start case should not use daemon, since this is specific
to Red Hat.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 867f57d166395c92949e480ca725249b0ca8950b)
Use a local variable $ctdbd so that we always run ctdbd from the the
same place and so that we know what to kill. This variable respects
the $CTDBD environment variable, which may be used to specify an
alternative location for the daemon.
In the important cases use "pkill -0 -f" to check if ctdbd is
running. Also, remove the special case for killing ctdbd when running
under valgrind. The regular case will handle this just fine.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 070305adfe636c2580776e6bf24bb8be06622b86)
Add a helper function that checks whether a unix domain socket exists
and there is a daemon LISTENING to it similar to the existing function
to check for a daemon LISTENING to a tcp/ip socket.
(This used to be ctdb commit 025a836ab3be3c078fccd8c10b10dfffbfdd94d0)
RHEL5 can SIGKILL httpd when stopping it, causing it to leak
semaphores. This means that eventually a node runs out of semaphores
and httpd can't be started. So, before we attempt to start httpd we
clean up any semaphores owned by apache. We also try to restart httpd
in the monitor event if httpd has gone away.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 2d3fbbbb63f443686f9fec42c0bc2058d115806e)
We want ctdb to shutdown first, as it manages many other
services. With the old level of 32 the NFS service would shutdown
first, and that would trigger ctdb to do a recovery. Then ctdb itself
would be shutdown a few seconds later, which causes a lot of error
messages in the other nodes logs
(This used to be ctdb commit 2f952af1a12e81a652ec9a4794db96f9593f2676)
This would allow a sysadmin to set up ctdb to send an email/snmptrap/... when the status of the node changes.
(This used to be ctdb commit ce534a83a05dbd40238e4eee0669d60ff396f935)
This can take very long if there are very many shares and is in that case better to implement in a separate cronjob than in ctdb eventscript
(This used to be ctdb commit 432604a1435cd2b5a7178fb5aedf1d4b61bffeb9)
this is a timeconsuming process and might not be feasible to perform if there are very many thousand shares
(This used to be ctdb commit 051ae5f3c13892b860818eac803d348f09845dc6)
The httpd service on suse and ubuntu/debian systems is usually
called "apache2" nowadays.
Note: There are older installs with Apache 1.3 out there, in which case
the service is called "apache". An extra check for these installs could
be useful as a sequel to this patch...
Michael
(This used to be ctdb commit b9e50e3416fecef6a881be3f1b91be977299293f)
The netstat test only grepped for the ipv4 wildcard address.
Now the ipv6 wildcard listener is correctly detected as well.
Michael
(This used to be ctdb commit 78e7928797e239e71f96eb001460a0dbf943e18f)
This fixes tcp port monitor events on systems, where netcat or nc
is not found in /usr/bin/, Debian, for instance.
The patch also separates the process of finding the binaries and
calling them, moving the detection outside of the loop over the
ports list.
Michael
(This used to be ctdb commit 3adf100e7f0c04aaf2da9ae4c6984cdb708c3b57)
