nivanova/samba-autobuild/.git
16 years agofixed a double close of a socket, leading to an EPOLL error
Andrew Tridgell [Mon, 22 Oct 2007 06:41:11 +0000 (16:41 +1000)]
fixed a double close of a socket, leading to an EPOLL error
(This used to be ctdb commit bbe8ad842bdfedd37ef14a6be07ad939113fe9b1)

16 years agoanother place where we need to mark connect_fde as freed
Andrew Tridgell [Mon, 22 Oct 2007 05:13:32 +0000 (15:13 +1000)]
another place where we need to mark connect_fde as freed
(This used to be ctdb commit d047fbeafebe4b150602f9a91802795659058b16)

16 years agofixed a valgrind uninitialised memory error due to pad bytes
Andrew Tridgell [Mon, 22 Oct 2007 05:13:08 +0000 (15:13 +1000)]
fixed a valgrind uninitialised memory error due to pad bytes
(This used to be ctdb commit aea9b0c8d467fe19815c046969e9c1049a3a20ac)

16 years agoprevent a double free
Andrew Tridgell [Mon, 22 Oct 2007 04:07:35 +0000 (14:07 +1000)]
prevent a double free
(This used to be ctdb commit 5a1b923abb36c6deb99ae178fdd54f12235dc309)

16 years agomerge from ronnie
Andrew Tridgell [Mon, 22 Oct 2007 00:26:25 +0000 (10:26 +1000)]
merge from ronnie
(This used to be ctdb commit b47fdc1fc86431c9159b595047faa76ba31f6829)

16 years agodont set parameters in statd-callout if they should be set they
Ronnie Sahlberg [Mon, 22 Oct 2007 00:18:38 +0000 (10:18 +1000)]
dont set parameters in statd-callout     if they should be set they
bshould be set from 10.interfaces

(This used to be ctdb commit 0c7c2dae0a976922de58793d576855bc37cd38e1)

16 years agodont set some of the sysctl variables in statd-callout. these are
Ronnie Sahlberg [Sat, 20 Oct 2007 20:42:33 +0000 (06:42 +1000)]
dont set some of the sysctl variables in statd-callout.    these are
mainly useful for avoiding ack-storms when doing very rapid
failover/failback during testing   but should not be required in
real-world.

this gets rid of a lof of annoying messages from the messages file

(This used to be ctdb commit 50d289dcce2caa7c7be9b6faa3b38b69c2237038)

16 years agomerge from tridge
Ronnie Sahlberg [Fri, 19 Oct 2007 05:19:25 +0000 (15:19 +1000)]
merge from tridge

(This used to be ctdb commit a45cfb29d9a0babccddc6aa26e71c00524da1d97)

16 years agoincrease release number
Andrew Tridgell [Fri, 19 Oct 2007 02:22:24 +0000 (12:22 +1000)]
increase release number
(This used to be ctdb commit 747ff96f1d93c52ba7548d0540266b0277d88ac1)

16 years agodont close the file, just set the fd to -1
Ronnie Sahlberg [Fri, 19 Oct 2007 01:03:12 +0000 (11:03 +1000)]
dont close the file, just set the fd to -1

(This used to be ctdb commit 04b26aa09e69b3c9fa1db245b5123c3cc02db8af)

16 years agomerge from ronnie
Andrew Tridgell [Thu, 18 Oct 2007 23:39:07 +0000 (09:39 +1000)]
merge from ronnie
(This used to be ctdb commit d444fdc7782496abe4b27003b647ac49fb52e6be)

16 years agoremove a incorrectly added file
Andrew Tridgell [Thu, 18 Oct 2007 23:30:55 +0000 (09:30 +1000)]
remove a incorrectly added file
(This used to be ctdb commit ff01a32db81b6c04d42634f5660181c270988264)

16 years agoadd missing ) in the IB transport (which i dont compile for)
Ronnie Sahlberg [Thu, 18 Oct 2007 23:05:37 +0000 (09:05 +1000)]
add  missing ) in the IB transport (which i dont compile for)

(This used to be ctdb commit 7f7a184bae87d46bd589d11068b6443b007366b4)

16 years agoadd a stub restart method for IB
Ronnie Sahlberg [Thu, 18 Oct 2007 23:04:52 +0000 (09:04 +1000)]
add a stub restart method for IB

(This used to be ctdb commit d318504ad5a49dbdfa307be39ae88df839e6308d)

16 years agoadd a new transport method so that when a node is marked as dead, we
Ronnie Sahlberg [Thu, 18 Oct 2007 22:58:30 +0000 (08:58 +1000)]
add a new transport method so that when a node is marked as dead, we
shut down and restart the transport

othervise, if we use the tcp transport the tcp connection might try to
retransmit the queued data during the time the node is unavailable.
this together with the exponential backoff for tcp means that the tcp
connection quickly reaches the maximum backoff rto which is often 60 or
120 seconds.   this would mean that it could take up to 60/120 seconds
before the tcp layer detects that the connection is dead and it has to
be reestablished.

(This used to be ctdb commit 0256db470879ce556b0f00070f7ebeaf37e529ab)

16 years agoset the flags explicitely isnstead of masking them in
Ronnie Sahlberg [Thu, 18 Oct 2007 06:54:00 +0000 (16:54 +1000)]
set the flags explicitely isnstead of masking them in

(This used to be ctdb commit 27a5f9dead44890683f9dbc4f07cda11264aa03b)

16 years agoadded some debug lines to help track down a problem
Andrew Tridgell [Thu, 18 Oct 2007 06:27:36 +0000 (16:27 +1000)]
added some debug lines to help track down a problem
(This used to be ctdb commit 2ca31e9de179f76e392a26cc8305e2473357c760)

16 years agomerge from tridge
Ronnie Sahlberg [Thu, 18 Oct 2007 05:53:50 +0000 (15:53 +1000)]
merge from tridge

(This used to be ctdb commit ad03e63906270c9c076ffdb1f62f912bb414ea10)

16 years agomerge from ronnie
Andrew Tridgell [Thu, 18 Oct 2007 05:51:15 +0000 (15:51 +1000)]
merge from ronnie
(This used to be ctdb commit a6b094fdede0ae850e87877fad0b9dd1f3a26869)

16 years agomerge from ronnie
Andrew Tridgell [Thu, 18 Oct 2007 05:44:02 +0000 (15:44 +1000)]
merge from ronnie
(This used to be ctdb commit 75d4b386293e186a6bb8532515585ab72670d663)

16 years agoflush the route cache when we have added the single public ip to the
Ronnie Sahlberg [Thu, 18 Oct 2007 04:13:48 +0000 (14:13 +1000)]
flush the route cache when we have added the single public ip to the
node

cleanup and remove everything when we do a shutdown event

(This used to be ctdb commit 221432f45073bc7624803058c8bbf18838e7ceeb)

16 years agouse NF_DROP instead of NF_STOLEN when we tell the kernel to not worry
Ronnie Sahlberg [Wed, 17 Oct 2007 05:03:58 +0000 (15:03 +1000)]
use NF_DROP instead of NF_STOLEN when we tell the kernel to not worry
about this packet any more and just forget it ever saw it

(This used to be ctdb commit 42a2a777cbc15a8cbbea7ecf2fb1c6dafa242d0c)

16 years agoreverse the order in which public ips are listed so it matches the order
Ronnie Sahlberg [Wed, 17 Oct 2007 03:42:42 +0000 (13:42 +1000)]
reverse the order in which public ips are listed so it matches the order
of the public_addresses file

(This used to be ctdb commit ce987661edd9160982e65866fb773445d296e5c7)

16 years agomerge from tridge
Ronnie Sahlberg [Wed, 17 Oct 2007 00:10:52 +0000 (10:10 +1000)]
merge from tridge

(This used to be ctdb commit 87760a95ec0a9e3cb2c415c569235a1ff58318cb)

16 years agoincrease release number
Andrew Tridgell [Tue, 16 Oct 2007 10:14:04 +0000 (20:14 +1000)]
increase release number
(This used to be ctdb commit 69fe7ce1d7874ce51d79de29adc53c207cb8869f)

16 years agomore detail on multipath config
Andrew Tridgell [Tue, 16 Oct 2007 10:13:28 +0000 (20:13 +1000)]
more detail on multipath config
(This used to be ctdb commit 78c44f2267cbef5fbc57d56dfd5ff40972733a1f)

16 years agoadd back the test inside the daemon that if someone asks us to drop
Ronnie Sahlberg [Tue, 16 Oct 2007 05:27:07 +0000 (15:27 +1000)]
add back the test inside the daemon that if someone asks us to drop
recovery mode back to NORMAL that we can not lock the reclock file
since at this stage it MUST be locked by the recovery daemon.

in order to avoid a non-blocking fnctl() lock from blocking and cause
"issues"  we move the 'test that we can not lock reclock file' into a
child process.

(This used to be ctdb commit 3af994641ec2234e37da1fa1f693441586471a7e)

16 years agoadd a new tunable : DeterministicIPs that makes the allocation of
Ronnie Sahlberg [Tue, 16 Oct 2007 02:15:02 +0000 (12:15 +1000)]
add a new tunable : DeterministicIPs  that makes the allocation of
public addresses to nodes deterministic.

Activate it by adding CTDB_SET_DeterministicIPs=1 in /etc/sysconfig/ctdb

When this is set,    the first entry in /etc/ctdb/public_addresses will
always be hosted by node 0, when that node is available, the second
entry by node1 and so on.

This tunable allows the allocation of addresses to become very
unbalanced and is only for debugging/testing use.
Beware, this feature requires that /etc/ctdb/public_addresses are
identical on all the nodes in the cluster.

(This used to be ctdb commit f0ca221f235731542090d8a6c86f2b7cd2ce2f96)

16 years agoinclude system/network.h so we get the prototype for inet_aton()
Ronnie Sahlberg [Tue, 16 Oct 2007 01:29:33 +0000 (11:29 +1000)]
include system/network.h so we get the prototype for inet_aton()

(This used to be ctdb commit 7145764b2d217f88a723dcb0ffd4e5a1567d64cf)

16 years agomerge from tridge
Ronnie Sahlberg [Tue, 16 Oct 2007 01:26:22 +0000 (11:26 +1000)]
merge from tridge

(This used to be ctdb commit 9e6bc12c9be2dabcfb9c6aeef257ef4737287fab)

16 years agodont try to lock the file from inside the ctdb daemon.
Ronnie Sahlberg [Mon, 15 Oct 2007 23:50:31 +0000 (09:50 +1000)]
dont try to lock the file from inside the ctdb daemon.
eventhough we dont want a blocking lock it does appear that the fcntl()
call can block for a while if gpfs is in the process of rebuilding
itself after a node arriving/leaving the cluster

(This used to be ctdb commit 6c0d206dea7116db71bccb4802a93dd7283249f6)

16 years agoonly link to -lipq if needed
Andrew Tridgell [Mon, 15 Oct 2007 04:44:06 +0000 (14:44 +1000)]
only link to -lipq if needed
(This used to be ctdb commit 7c378d881e37db0f14e07ccba19fde1f9f4f0831)

16 years agoimproved handling of systems without libipq.h
Andrew Tridgell [Mon, 15 Oct 2007 04:37:54 +0000 (14:37 +1000)]
improved handling of systems without libipq.h
(This used to be ctdb commit cfa8ddd3ca53c0160558137cccfc7e73e46ec36c)

16 years agodisable ipmux code until we have a configure test
Andrew Tridgell [Mon, 15 Oct 2007 04:29:47 +0000 (14:29 +1000)]
disable ipmux code until we have a configure test
(This used to be ctdb commit fd83f0f3eb233f22ce9b5b4afbc4f26e3c865b3c)

16 years agosync flags between nodes in monitor loop in recmaster
Andrew Tridgell [Mon, 15 Oct 2007 04:28:51 +0000 (14:28 +1000)]
sync flags between nodes in monitor loop in recmaster
(This used to be ctdb commit 6eef86e06388fc53a1212f1e2783ae174c6cd210)

16 years agomerge from ronnie
Andrew Tridgell [Mon, 15 Oct 2007 04:17:49 +0000 (14:17 +1000)]
merge from ronnie
(This used to be ctdb commit d18712caba11855010be52f90bac656683076676)

16 years agodisable optimisation for now, until we find a occasional segv
Andrew Tridgell [Mon, 15 Oct 2007 03:31:09 +0000 (13:31 +1000)]
disable optimisation for now, until we find a occasional segv
(This used to be ctdb commit d09570c70551aa40390ce9ceffe7bc234e1afafe)

16 years agoadd config option for disabling bans
Andrew Tridgell [Mon, 15 Oct 2007 03:22:58 +0000 (13:22 +1000)]
add config option for disabling bans
(This used to be ctdb commit 153b911f7f957d4c564b04f5aa878033a02da9e4)

16 years agouse $CTDB_BASE in 90.ipmux instead of hardcoding it to /etc/ctdb
Ronnie Sahlberg [Wed, 10 Oct 2007 21:51:57 +0000 (07:51 +1000)]
use $CTDB_BASE in 90.ipmux instead of hardcoding it to /etc/ctdb

(This used to be ctdb commit 6abb46b010851f5719f12273b4a3d46ec986f0c7)

16 years agouse kill_tcp_connections() to kill off all tcp connections to the
Ronnie Sahlberg [Wed, 10 Oct 2007 21:30:10 +0000 (07:30 +1000)]
use kill_tcp_connections() to kill off all tcp connections to the
"single public ip" address when we do a recovery

(This used to be ctdb commit 19b52a2d5db31efa9e7c77037097ff8539986ac3)

16 years agomove the kill_tcp_connections() function from 10.interfaces to functions
Ronnie Sahlberg [Wed, 10 Oct 2007 21:27:38 +0000 (07:27 +1000)]
move the kill_tcp_connections() function from 10.interfaces to functions

(This used to be ctdb commit 055948530fb16bf49c42fc4489f29a21665156c0)

16 years agofirst check that recovery master is connected (we know this from our own
Ronnie Sahlberg [Wed, 10 Oct 2007 21:10:17 +0000 (07:10 +1000)]
first check that recovery master is connected (we know this from our own
flags)

then pull the flags off recovery master before checking if it is banned

(This used to be ctdb commit 94c1d234e57a40eda2d8b892dd9fbe1ffc4b3433)

16 years agosimplify election handling
Ronnie Sahlberg [Wed, 10 Oct 2007 20:16:36 +0000 (06:16 +1000)]
simplify election handling

make sure we read and update the flags from all remote nodes before we
reach the first codepath that can call do_recovery()
since during do_recovery() we need to know what the flags are.

(This used to be ctdb commit e85f3806483ea420559d449e0e4d81bec996740f)

16 years agomerge from tridge
Ronnie Sahlberg [Wed, 10 Oct 2007 00:49:55 +0000 (10:49 +1000)]
merge from tridge

(This used to be ctdb commit 4690a205fe4325b03ab044bdb5fbc9aa3e94db6e)

16 years agomake sure reconnected nodes start off as unhealthy so they don't get a public IP
Andrew Tridgell [Wed, 10 Oct 2007 00:45:22 +0000 (10:45 +1000)]
make sure reconnected nodes start off as unhealthy so they don't get a public IP
(This used to be ctdb commit c733ec6760cae01ce277f491caf1355e46de5cf7)

16 years agoadd a --single-public-ip argument to ctdbd to specify the ip address
Ronnie Sahlberg [Tue, 9 Oct 2007 23:42:32 +0000 (09:42 +1000)]
add a --single-public-ip argument to ctdbd to specify the ip address
used in single public ip address mode.
when using this argument, --public-interface must also be used.

add a vnn structure to the ctdb context to describe the single public ip
address

update the killtcp control in the daemon that if a socketpair that is to
be killed does not match a normal public address it checks if the
destination address maches the single public ip address and if so uses
that vnn structure from the ctdb context

this allows killtcp to kill also connections to the single public ip
instead of only normal public addresses

(This used to be ctdb commit 5661ba17b91f62821dec1c76056c78b99752a90b)

16 years agoremove some debug outputs
Ronnie Sahlberg [Tue, 9 Oct 2007 03:45:42 +0000 (13:45 +1000)]
remove some debug outputs

(This used to be ctdb commit f29c0b52df1f455909ba133e3ad3bc462dc32929)

16 years agosend out gratious arps when we are starting up serving the "single
Ronnie Sahlberg [Tue, 9 Oct 2007 02:00:12 +0000 (12:00 +1000)]
send out gratious arps when we are starting up serving the "single
public ip" but before we start the ipmux tool

(This used to be ctdb commit dad1a80f39763314825939095f7656c13dcdbdc3)

16 years agoadd a control to send gratious arps from the ctdb daemon
Ronnie Sahlberg [Tue, 9 Oct 2007 01:56:09 +0000 (11:56 +1000)]
add a control to send gratious arps from the ctdb daemon

(This used to be ctdb commit 563819dd1acb344f95aabb4bad990b36f7ea4520)

16 years agoadd an initial test version of an ip multiplex tool that allows us
Ronnie Sahlberg [Mon, 8 Oct 2007 04:05:22 +0000 (14:05 +1000)]
add an initial test version of an ip multiplex tool that allows us
to have one single public ip address for the entire cluster.

this ip address is attached to lo on all nodes but only the recmaster
will respond to arp requests for this address.
the recmaster then runs an ipmux process that will pass any incoming
packets to this ip address onto the other node sin the cluster based on
the ip address of the client host

to use this feature one must
1, have one fixed ip address in the customers network attached
permanently attached to an interface
2, set CTDB_PUBLI_INTERFACE=
   to specify on which interface the clients attach to the node
3, CTDB_SINGLE_PUBLI_IP=ip-address
   to specify which ipaddress should be the "single public ip address"

to test with only one single client,   attach several ip addresses to
the client and ping the public address from the client with different -I
options.   look in network trace to see to which node the packet is
passed onto.

(This used to be ctdb commit 50d648c95e4e6d7c2867a034c2b550086d853320)

16 years agoadd a function in the ctdb tool to determine whether the local node is
Ronnie Sahlberg [Sun, 7 Oct 2007 23:47:20 +0000 (09:47 +1000)]
add a function in the ctdb tool to determine whether the local node is
the recmaster or not.

return 0 if the node is the recmaster and 1 (true) if it is not or if
we could not communicate with the ctdb daemon.

call it 'isnotrecmaster' to cope with that if the tool could not bind to
the socket to tyalk to the daemon, the tool will automatically return an
error and exit code 1
thus the tool will only return 0 if it could talk successfully to the
local daemon and if the local daemon confirms this node is the recmaster

(This used to be ctdb commit ae5fcb790b6c3985f514fa8a96bc00c2619f2a28)

16 years agomerge from tridge
Ronnie Sahlberg [Fri, 5 Oct 2007 22:11:24 +0000 (08:11 +1000)]
merge from tridge

(This used to be ctdb commit 02cda01c032804cb1c53593ceb98685c827e2d58)

16 years agofixed several places where we set the recovery culprit incorrectly
Andrew Tridgell [Fri, 5 Oct 2007 03:51:31 +0000 (13:51 +1000)]
fixed several places where we set the recovery culprit incorrectly
(This used to be ctdb commit d9da73395fa443801fc68ec53a42b548e832d58a)

16 years ago - catch ESTALE in the recovery lock by trying a read()
Andrew Tridgell [Fri, 5 Oct 2007 03:28:21 +0000 (13:28 +1000)]
 - catch ESTALE in the recovery lock by trying a read()
- priortise nodes that are unbanned and healthy in the election

(This used to be ctdb commit 929feb475dfdf7283f0e99b50b179e1c91d3a39f)

16 years agowe are the culprit if we can't get the reclock
Andrew Tridgell [Fri, 5 Oct 2007 02:01:40 +0000 (12:01 +1000)]
we are the culprit if we can't get the reclock
(This used to be ctdb commit 1d320e113c6134ff6822b985a47131d8204af35a)

16 years agochange async.private to async.private_data since private is a reserved
Ronnie Sahlberg [Wed, 26 Sep 2007 04:25:32 +0000 (14:25 +1000)]
change async.private to async.private_data since private is a reserved
work in c++

(This used to be ctdb commit 79eb28f6cd5dcc30b04966d202a050eaf98a2552)

17 years agomerge from tridge
Ronnie Sahlberg [Tue, 25 Sep 2007 01:43:42 +0000 (11:43 +1000)]
merge from tridge

(This used to be ctdb commit 5655fab1284dce8f4a09ad426d53f5151c88968b)

17 years agoupped version number
Andrew Tridgell [Mon, 24 Sep 2007 05:27:01 +0000 (15:27 +1000)]
upped version number
(This used to be ctdb commit 4312e20e047ddb0f825c5e0c51d85dfa6a1b7df8)

17 years agomerge from ronnie
Andrew Tridgell [Mon, 24 Sep 2007 03:52:35 +0000 (13:52 +1000)]
merge from ronnie
(This used to be ctdb commit c67f516f01f8033e3fbd0f338eaa3a8afb862495)

17 years agowhen we have a public ip address mismatch (i.e. we hold addresses we
Ronnie Sahlberg [Mon, 24 Sep 2007 00:52:26 +0000 (10:52 +1000)]
when we have a public ip address mismatch (i.e. we hold addresses we
shouldnt   or we are not holding addresses wqe should)
we must first freeze the local node before we set the recovery mode

(This used to be ctdb commit a77a77e8b5180f6a4a1f3d7d4ff03811f3b71b56)

17 years agomerge from tridge
Ronnie Sahlberg [Mon, 24 Sep 2007 00:27:48 +0000 (10:27 +1000)]
merge from tridge

(This used to be ctdb commit 7f9242747543ea1a2cc05f5c8afc51ab26e7d4bb)

17 years agofixed a fd leak on the recovery lock
Andrew Tridgell [Mon, 24 Sep 2007 00:19:07 +0000 (10:19 +1000)]
fixed a fd leak on the recovery lock
(This used to be ctdb commit 186f35c42ed4fcc9ed44390b0dd036ece475d45e)

17 years agorun monitoring more quickly when unhealthy and at startup
Andrew Tridgell [Mon, 24 Sep 2007 00:12:18 +0000 (10:12 +1000)]
run monitoring more quickly when unhealthy and at startup
(This used to be ctdb commit ff1c205928e3ef5bcc6bf4e4b2122a19fa38d8f4)

17 years agono longer wait at startup for services to become available, instead
Andrew Tridgell [Mon, 24 Sep 2007 00:00:14 +0000 (10:00 +1000)]
no longer wait at startup for services to become available, instead
set the node initially unhealthy and let the status monitoring bring the node online.
This fixes a problem with winbindd, where it refused to start because secrets.tdb was not populated
but we could not populate ctdbd, because the net command would not run while ctdbd was still doing startup
and thus frozen
(This used to be ctdb commit 3a001b793dd76fb96addf1e2ccb74da326fbcfbc)

17 years agofixed a valgrind error, and some warnings
Andrew Tridgell [Sun, 23 Sep 2007 23:57:14 +0000 (09:57 +1000)]
fixed a valgrind error, and some warnings
(This used to be ctdb commit c0f52dbb385fa0748680adb7c40755c92e577551)

17 years agomake the persistent dbdir configurable
Andrew Tridgell [Fri, 21 Sep 2007 06:12:04 +0000 (16:12 +1000)]
make the persistent dbdir configurable
(This used to be ctdb commit 2587b887dcfce26b12c66fcb5d34e92da42a1776)

17 years agomerge from tridge
Ronnie Sahlberg [Fri, 21 Sep 2007 05:45:48 +0000 (15:45 +1000)]
merge from tridge

(This used to be ctdb commit 01c388f12ed554bda1a46b21cd18bf1b00f962c3)

17 years agoavoid using connected nodes that aren't in the vnn map yet
Andrew Tridgell [Fri, 21 Sep 2007 05:44:13 +0000 (15:44 +1000)]
avoid using connected nodes that aren't in the vnn map yet
(This used to be ctdb commit 2b5ae133f5f6fa9ad1a8896fe4b4c542d4ca462d)

17 years agomerge bugfix from ronnie
Andrew Tridgell [Fri, 21 Sep 2007 05:32:11 +0000 (15:32 +1000)]
merge bugfix from ronnie
(This used to be ctdb commit c179b0085ae239b576139545c43f1c141a03e196)

17 years agoin ctdb_control_persistent_store() we must talloc_steal() the pointer to
Ronnie Sahlberg [Fri, 21 Sep 2007 05:19:33 +0000 (15:19 +1000)]
in ctdb_control_persistent_store() we must talloc_steal() the pointer to
c   to prevent it from being immediately freed (and our persistent store
state with it) if we need to wait asynchronously for other nodes before
we can reply back to the client

(This used to be ctdb commit fa5915280933e4d2e7d4d07199829c9c2b87a335)

17 years agomerge from ronnie
Andrew Tridgell [Fri, 21 Sep 2007 04:47:32 +0000 (14:47 +1000)]
merge from ronnie
(This used to be ctdb commit 485182ba7b91b756f93be4863bf5fa208d460a54)

17 years agowhen ctdb attaches to a database it broadcasts the attach to all other
Ronnie Sahlberg [Fri, 21 Sep 2007 03:47:40 +0000 (13:47 +1000)]
when ctdb attaches to a database  it broadcasts the attach to all other
nodes so that the db is created on them as well

when we send this broadcast   we must use the correct control and not
assume all databases created are of the temporary kind

(This used to be ctdb commit 106f816d4a0814ca4418de051289d9fc62df7dd2)

17 years agomerge from tridge
Ronnie Sahlberg [Fri, 21 Sep 2007 03:20:29 +0000 (13:20 +1000)]
merge from tridge

(This used to be ctdb commit bb283ee8ebaea848366e9c3b3d3244da459a7967)

17 years agoadded support for persistent databases in ctdbd
Andrew Tridgell [Fri, 21 Sep 2007 02:24:02 +0000 (12:24 +1000)]
added support for persistent databases in ctdbd
(This used to be ctdb commit 3115090a0d882beca9d70761130b74bb0821f201)

17 years agomerge from tridge
Ronnie Sahlberg [Wed, 19 Sep 2007 01:54:45 +0000 (11:54 +1000)]
merge from tridge

(This used to be ctdb commit c4262a3f980a9fabe0985fc3281e9f5de1cdf35e)

17 years agoone more command to run to enable winbind for vsftpd
Ronnie Sahlberg [Wed, 19 Sep 2007 01:53:48 +0000 (11:53 +1000)]
one more command to run to enable winbind for vsftpd

(This used to be ctdb commit 1f6d13a364cde58b66a4bc52e909cc68b8c807d7)

17 years agomake sure we set close on exec on any possibly inherited fds
Andrew Tridgell [Wed, 19 Sep 2007 01:46:37 +0000 (11:46 +1000)]
make sure we set close on exec on any possibly inherited fds
(This used to be ctdb commit d9dec82076f14a348e7b67b4350180681ff86f32)

17 years agoseparate out the various fs display ops
Andrew Tridgell [Wed, 19 Sep 2007 01:46:11 +0000 (11:46 +1000)]
separate out the various fs display ops
(This used to be ctdb commit dc89e1a428da5d5ca2a9c4988c05de3ea65f00f4)

17 years agoexpanded ctdb_diagnostics a bit
Andrew Tridgell [Mon, 17 Sep 2007 05:31:33 +0000 (15:31 +1000)]
expanded ctdb_diagnostics a bit
(This used to be ctdb commit 70a4bb3dc7e624ad778949dbc874c2617fd532e6)

17 years agoadd documantation of additional requirements for FTP so that users can
Ronnie Sahlberg [Mon, 17 Sep 2007 03:01:16 +0000 (13:01 +1000)]
add documantation of additional requirements for FTP so that users can
log in and access files using the AD username/password

(This used to be ctdb commit 679e125770247fc24dfb14b5781d44f639457ecd)

17 years agomerge from tridge
Ronnie Sahlberg [Sun, 16 Sep 2007 21:43:15 +0000 (07:43 +1000)]
merge from tridge

(This used to be ctdb commit 6aa62f1b42c151e9a5760d927be484ce8a8540b0)

17 years agoincrease release number
Andrew Tridgell [Fri, 14 Sep 2007 09:27:11 +0000 (19:27 +1000)]
increase release number
(This used to be ctdb commit b213f4f1bf5bcfb40b9ce176df22216e3ebbe964)

17 years agomerge from ronnie
Andrew Tridgell [Fri, 14 Sep 2007 05:23:23 +0000 (15:23 +1000)]
merge from ronnie
(This used to be ctdb commit 913c33a7d2f67570548fecc568dba874e5f72dd2)

17 years agolet ctdb ip only print the ip addresses known to the specified node
Ronnie Sahlberg [Fri, 14 Sep 2007 05:19:44 +0000 (15:19 +1000)]
let ctdb ip   only print the ip addresses known to the specified node
and not the entire cluster

(This used to be ctdb commit eb1f67a56d752c9f42a9a26a6697a7ab8e668b3a)

17 years agoupdate vnn -> pnn in documentation
Ronnie Sahlberg [Fri, 14 Sep 2007 04:24:53 +0000 (14:24 +1000)]
update vnn -> pnn in documentation

(This used to be ctdb commit bb62b4df514255b95d3871b254bec9c440bc4a06)

17 years agodocumentation updates
Ronnie Sahlberg [Fri, 14 Sep 2007 04:19:12 +0000 (14:19 +1000)]
documentation updates

it is --event-script-dir      not --event-script

add explanation of the public_addresses file

(This used to be ctdb commit 21325b23e786ac1c2abc07ea75b0814e9c725a9e)

17 years agocope with non-standard install dirs in event scripts
Andrew Tridgell [Fri, 14 Sep 2007 04:14:03 +0000 (14:14 +1000)]
cope with non-standard install dirs in event scripts
(This used to be ctdb commit 52fff5345873690a9cc86495f414343eaa3bd540)

17 years agomerge from tridge
Ronnie Sahlberg [Fri, 14 Sep 2007 02:18:34 +0000 (12:18 +1000)]
merge from tridge

(This used to be ctdb commit 340b4a0a31a50385430fea015b146c84d0b2bbb2)

17 years agofix pkill args
Andrew Tridgell [Fri, 14 Sep 2007 01:59:04 +0000 (11:59 +1000)]
fix pkill args
(This used to be ctdb commit 9690de97b4746f4a79830465e3a1679e9fbda671)

17 years agomake sure all public IPs are removed at startup
Andrew Tridgell [Fri, 14 Sep 2007 01:56:40 +0000 (11:56 +1000)]
make sure all public IPs are removed at startup
(This used to be ctdb commit b16f33787f2a9471285037f4a6d470e826536570)

17 years agoduring startup make sure to delete any public addresses from any
Ronnie Sahlberg [Fri, 14 Sep 2007 00:37:10 +0000 (10:37 +1000)]
during startup make sure to delete any public addresses from any
interface

(This used to be ctdb commit 18d80ea6db39e61f60e4c01de164d58bcbd8ab10)

17 years agolet each node verify that they have a correct assignment of public ip
Ronnie Sahlberg [Fri, 14 Sep 2007 00:16:36 +0000 (10:16 +1000)]
let each node verify that they have a correct assignment of public ip
addresses (i.e. htey hold those they should hold   and they dont hold
any of those they shouldnt hold)

if an inconsistency is found, mark the local node as recovery mode
active
and wait for the recovery master to trigger a full blown recovery

(This used to be ctdb commit 55a5bfc8244c5b9cdda3f11992f384f00566b5dc)

17 years ago- merge from ronnie
Andrew Tridgell [Thu, 13 Sep 2007 23:49:12 +0000 (09:49 +1000)]
- merge from ronnie
- add a flag to check that recovery completed correctly. If not, re-trigger it in monitoring

(This used to be ctdb commit d5ed941d9bab4af30d8b5f9b77bdf43d9218d69b)

17 years agowait for ctdbd to finish cleanup before considering "service ctdb stop" to be done
Andrew Tridgell [Thu, 13 Sep 2007 23:25:11 +0000 (09:25 +1000)]
wait for ctdbd to finish cleanup before considering "service ctdb stop" to be done
(This used to be ctdb commit 216eb4be7ec481cfe9aaeeada257b77cb394d2e4)

17 years agonicer use of testparm
Andrew Tridgell [Thu, 13 Sep 2007 23:24:34 +0000 (09:24 +1000)]
nicer use of testparm
(This used to be ctdb commit a611ea930fb9dae6e56f6a74b2bdc9e08066d4d1)

17 years agoupdate the section about event scripts
Ronnie Sahlberg [Thu, 13 Sep 2007 22:56:27 +0000 (08:56 +1000)]
update the section about event scripts

(This used to be ctdb commit a0744480c85a4e8648bd0ae7600f90d311b931ea)

17 years agodisable nfsv4 in etc/sysconfig/nfs
Ronnie Sahlberg [Thu, 13 Sep 2007 22:15:24 +0000 (08:15 +1000)]
disable nfsv4 in etc/sysconfig/nfs

(This used to be ctdb commit b71e11f0e27bb3ff908ad171aa5b1f724609ad05)

17 years agowhen a ctdb_takeover_run has failed we must make sure that
Ronnie Sahlberg [Thu, 13 Sep 2007 04:51:37 +0000 (14:51 +1000)]
when a ctdb_takeover_run has failed  we must make sure that
need_takeover_run is set to true  or else we might forget to rerun it
again during the next recovery

othervise,  need_takeover_run is only set to true IFF the node flags for
a remote node and the local nodes differ.
It is possible that a takeover run fails  and thus the reassignment of
ip addresses is incomplete  but before we get back to the test in
monitor_cluster()  that all the node flags of all nodes have converged
and they now match each others again.   and thus causing
monitor_cluster() to fail to realize that a takeover run is needed.

(This used to be ctdb commit ae7e866787cebd14394983ce1834387c959d1022)

17 years agoensure smbd and winbindd do die in 50.samba
Andrew Tridgell [Thu, 13 Sep 2007 04:36:23 +0000 (14:36 +1000)]
ensure smbd and winbindd do die in 50.samba
(This used to be ctdb commit 6f23affedb626fc7a5ca86c4763f3045a5586231)

17 years agomerge from tridge
Ronnie Sahlberg [Thu, 13 Sep 2007 04:28:18 +0000 (14:28 +1000)]
merge from tridge

(This used to be ctdb commit eda3caa77be352967a41ff9bddda5296c94797a9)