Bug 1158805
| Summary: | Add support for qdevice/qnetd provided by corosync | ||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | Red Hat Enterprise Linux 7 | Reporter: | Tomas Jelinek <tojeline> | ||||||||||||||||||||||||||||||||
| Component: | pcs | Assignee: | Tomas Jelinek <tojeline> | ||||||||||||||||||||||||||||||||
| Status: | CLOSED ERRATA | QA Contact: | cluster-qe <cluster-qe> | ||||||||||||||||||||||||||||||||
| Severity: | medium | Docs Contact: | Steven J. Levine <slevine> | ||||||||||||||||||||||||||||||||
| Priority: | high | ||||||||||||||||||||||||||||||||||
| Version: | 7.1 | CC: | cfeist, cluster-maint, idevat, jpokorny, mlisik, mnavrati, rbednar, rsteiger, tlavigne, tojeline | ||||||||||||||||||||||||||||||||
| Target Milestone: | rc | Keywords: | FutureFeature, TechPreview | ||||||||||||||||||||||||||||||||
| Target Release: | --- | ||||||||||||||||||||||||||||||||||
| Hardware: | Unspecified | ||||||||||||||||||||||||||||||||||
| OS: | Unspecified | ||||||||||||||||||||||||||||||||||
| Whiteboard: | |||||||||||||||||||||||||||||||||||
| Fixed In Version: | pcs-0.9.152-10.el7 | Doc Type: | Enhancement | ||||||||||||||||||||||||||||||||
| Doc Text: |
Support for quorum devices in a Pacemaker cluster
Red Hat Enterprise Linux 7.4 provides full support for quorum devices, previously available as a Technology Preview. This feature provides the ability to configure a separate quorum device (QDevice) which acts as a third-party arbitration device for the cluster. Its primary use is to allow a cluster to sustain more node failures than standard quorum rules allow. A quorum device is recommended for clusters with an even number of nodes and highly recommended for two-node clusters. For information on configuring a quorum device, see https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/html/High_Availability_Add-On_Reference/.
|
Story Points: | --- | ||||||||||||||||||||||||||||||||
| Clone Of: | Environment: | ||||||||||||||||||||||||||||||||||
| Last Closed: | 2016-11-03 20:53:29 UTC | Type: | Bug | ||||||||||||||||||||||||||||||||
| Regression: | --- | Mount Type: | --- | ||||||||||||||||||||||||||||||||
| Documentation: | --- | CRM: | |||||||||||||||||||||||||||||||||
| Verified Versions: | Category: | --- | |||||||||||||||||||||||||||||||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||||||||||||||||||||||||||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||||||||||||||||||||||||||||||
| Embargoed: | |||||||||||||||||||||||||||||||||||
| Bug Depends On: | 614122 | ||||||||||||||||||||||||||||||||||
| Bug Blocks: | 1296673, 1313485, 1376485, 1376495 | ||||||||||||||||||||||||||||||||||
| Attachments: |
|
||||||||||||||||||||||||||||||||||
|
Description
Tomas Jelinek
2014-10-30 09:41:37 UTC
Created attachment 1146010 [details] proposed fix 1 Added command for updating quorum options (WFA, ATB, LMS), upstream patches: https://github.com/feist/pcs/commit/3d3ab9e512645a4b5962fc89c2a17dd61b0d2200 https://github.com/feist/pcs/commit/f3b5bd1fccd131af5e4c533ada4a31f2321ff621 https://github.com/feist/pcs/commit/254c57660c070fb8290d768e95cd18cfb5f28032 https://github.com/feist/pcs/commit/6b76c658bf5453ca4ed70353899f1296a8a9951c https://github.com/feist/pcs/commit/2d70bfb8fa63c44ef8181023941f34967af25a70 https://github.com/feist/pcs/commit/b28f9290b909fe758f90e23980fc76fb0ac1602d https://github.com/feist/pcs/commit/4a799a8fa46cf01fa61c45a57d00f47889799476 Created attachment 1147667 [details] proposed fix 2 automatically set two_node when user sets auto_tie_breaker upstream patch: https://github.com/feist/pcs/commit/b8a4beb9ddad151f900ddfea93e7620b7fcf2da2 Created attachment 1147685 [details] proposed fix 3 Add command for displaying quorum configuration, upstream patches: https://github.com/feist/pcs/commit/f33e1a01069b7974f6fca76129a9003a7c05dc77 https://github.com/feist/pcs/commit/1af7a3ab477136c6a90214d2bcd69979e3b84686 Created attachment 1152229 [details] proposed fix 4 Add preliminary support for add, remove and update qdevice in a cluster, upstream patches: https://github.com/feist/pcs/commit/7f3092a9dc99ef01a4b96373f3aca37188e06010 https://github.com/feist/pcs/commit/3a50c7999be33cbe73b94ed0049b7c0fc296b995 https://github.com/feist/pcs/commit/fcb2491c0ebc5597355ba1f41717cc3d89679cfb https://github.com/feist/pcs/commit/e348ea84ff4ddd5ff153103a3364f42863ff4bca Created attachment 1155252 [details] proposed fix 5 validate qdevice options, upstream patches: https://github.com/feist/pcs/commit/76ec4028cb726edcd74466bb5d5bdc2c15f608b7 https://github.com/feist/pcs/commit/63791c4b3d5983111c64c7f19c1add874d5defcd https://github.com/feist/pcs/commit/dba1976368479d3cfc8fa51b9de24bfdd4db4027 https://github.com/feist/pcs/commit/53492c271b7d15174016a6002b4366f68307d23d https://github.com/feist/pcs/commit/ec47ce93689ff82c596d04d3ca17e31cab67a2bc Created attachment 1156185 [details] proposed fix 6 allow to use --force flag in quorum commands to proceed in case of noncritical errors, upstream patches: https://github.com/feist/pcs/commit/d451498bc931a73ef52835e90dac7d269f397617 https://github.com/feist/pcs/commit/5b65d29c3d6d034e712418474f82cdf7f2c48a47 https://github.com/feist/pcs/commit/a2c2bb0fbb8e2744f6556282a53d0a14740641c2 https://github.com/feist/pcs/commit/aaaf95e50249f27bfbc21e8564569b271b6677b0 https://github.com/feist/pcs/commit/d4c04d68da3167770710415b63efd67ae5f7abf3 Created attachment 1159974 [details] proposed fix 7 check that corosync is stopped when configuring quorum, allow to skip offline nodes when synchronizing corosync configuration in quorum commands; upstream patches: https://github.com/feist/pcs/commit/22c6443636f07bc8845261620c268062cfc82845 https://github.com/feist/pcs/commit/869f3121ce7017e11e72a90f63e39147388bc060 https://github.com/feist/pcs/commit/816fc8d405ecc9249f5f1a5974f030091e9e28ae Created attachment 1162792 [details] proposed fix 8 Handle lms - 2nodelms algorithm changes when adding / removing nodes; upstream patches: https://github.com/feist/pcs/commit/d2a0fe41b7c1d02b1b803dcfbd31fad29349de3a https://github.com/feist/pcs/commit/f6525cd5425e445cb3d0f8965243df68852eb564 https://github.com/feist/pcs/commit/9c60d73dcf0b914c4e9d99d1d08aa4ddf8b1afe5 Created attachment 1162793 [details] proposed fix 9 Add qnetd service management; upstream patches: https://github.com/feist/pcs/commit/edfa10c20c915daab747ad0dcb5859f92dfe09c1 https://github.com/feist/pcs/commit/c8ce0efba11667ca4b8c86bfc8642e7a0e993b15 https://github.com/feist/pcs/commit/8c3adbfad892099387106102e643669cc2fcafdb https://github.com/feist/pcs/commit/c1bd824c981c3d84fd31f8999ddce57221ae357c https://github.com/feist/pcs/commit/b8fe22ecf995619c4efbfcd12008abb0e2e50256 Created attachment 1173889 [details] proposed fix 10 full qdevice support in cli, upstream patches: https://github.com/ClusterLabs/pcs/commit/992090ff1ee39002df375ffd4bd91e5f586399b3 https://github.com/ClusterLabs/pcs/commit/6188e98c17839f0fef8ec53f64b36d4721fb9dae https://github.com/ClusterLabs/pcs/commit/ae9fc132db04777974b180efbef0e779c46a689f https://github.com/ClusterLabs/pcs/commit/2d16bbbe1bfb25cb89d6a2a11a6494222b8ff1ad https://github.com/ClusterLabs/pcs/commit/c19cbdc0cef46f6724adfe33db355c8b265a49e6 https://github.com/ClusterLabs/pcs/commit/d3671294a211d5f4611d49da0b976497ae8bf266 https://github.com/ClusterLabs/pcs/commit/0bb9d53b7e972fce0c94ed9106daed8ae372b548 https://github.com/ClusterLabs/pcs/commit/152f76eb2e59c108a25aba35969bca69debd80df https://github.com/ClusterLabs/pcs/commit/f44f1c2c324e432b9cd217c3678a05d570a6aee6 https://github.com/ClusterLabs/pcs/commit/60524ed0f2b94fe4ab0c08d8a2f464b9b3598096 https://github.com/ClusterLabs/pcs/commit/3418ad8168abecab2fb6b4f2a614bec2de85a25d https://github.com/ClusterLabs/pcs/commit/53e72d6980a9f3f69b052f7c196d463460a27e87 https://github.com/ClusterLabs/pcs/commit/f55854b1bdf7e4984dba0b659ee985325d3c0cfa https://github.com/ClusterLabs/pcs/commit/bb5763a4c9a31f5fcd33cd2b96257a9aba0c32c8 https://github.com/ClusterLabs/pcs/commit/ad3114930258ee9b29211f10aa5184e4aa69e1bb https://github.com/ClusterLabs/pcs/commit/a3af280303e6918b0ce664c4691ff11365c3b33e https://github.com/ClusterLabs/pcs/commit/2ac9f689dec06923b4f6e2fbf8cca65f7b1d2299 https://github.com/ClusterLabs/pcs/commit/5673a2d9bee5dfbd2f2e6c3faff5262b40aadb77 https://github.com/ClusterLabs/pcs/commit/e1019d761e33afddce5931f81457551a7a5d7713 https://github.com/ClusterLabs/pcs/commit/bf531dea9af0fe48c3aee84139d2434a5be51b7a https://github.com/ClusterLabs/pcs/commit/d226c4398d2e25e1883813902ff49ed4fd30758c https://github.com/ClusterLabs/pcs/commit/77be353fe5199a1829987865e7bded13f2974979 Usage and test: > requirements: - corosync and corosync-qdevice installed on cluster nodes - corosync-qnetd installed on a quorum device host > First we need to setup the quorum device host: [root@rh72-node3:~]# pcs qdevice setup model net --enable --start Quorum device 'net' initialized quorum device enabled Starting quorum device... quorum device started > We can see qnetd is running and there are no clients connected to it: [root@rh72-node3:~]# pcs qdevice status net --full QNetd address: *:5403 TLS: Supported (client certificate required) Connected clients: 0 Connected clusters: 0 Maximum send/receive size: 32768/32768 bytes > Pcs provides standard set of commands for managing qnetd service: pcs qdevice start|stop|enable|disable|kill net > Let's add our new quorum device to a cluster. > We can see the cluster is not using a quorum device yet: [root@rh72-node1:~]# pcs quorum config Options: [root@rh72-node1:~]# pcs quorum status Quorum information ------------------ Date: Wed Jun 29 13:15:36 2016 Quorum provider: corosync_votequorum Nodes: 2 Node ID: 1 Ring ID: 1/8272 Quorate: Yes Votequorum information ---------------------- Expected votes: 2 Highest expected: 2 Total votes: 2 Quorum: 1 Flags: 2Node Quorate Membership information ---------------------- Nodeid Votes Qdevice Name 1 1 NR rh72-node1 (local) 2 1 NR rh72-node2 > Let's add our quorum device then: [root@rh72-node1:~]# pcs quorum device add model net host=rh72-node3 algorithm=lms Setting up qdevice certificates on nodes... rh72-node2: Succeeded rh72-node1: Succeeded Enabling corosync-qdevice... rh72-node1: corosync-qdevice enabled rh72-node2: corosync-qdevice enabled Sending updated corosync.conf to nodes... rh72-node1: Succeeded rh72-node2: Succeeded Corosync configuration reloaded Starting corosync-qdevice... rh72-node1: corosync-qdevice started rh72-node2: corosync-qdevice started > Check it from the cluster side: [root@rh72-node1:~]# pcs quorum config Options: Device: Model: net algorithm: lms host: rh72-node3 [root@rh72-node1:~]# pcs quorum status Quorum information ------------------ Date: Wed Jun 29 13:17:02 2016 Quorum provider: corosync_votequorum Nodes: 2 Node ID: 1 Ring ID: 1/8272 Quorate: Yes Votequorum information ---------------------- Expected votes: 3 Highest expected: 3 Total votes: 3 Quorum: 2 Flags: Quorate Qdevice Membership information ---------------------- Nodeid Votes Qdevice Name 1 1 A,V,NMW rh72-node1 (local) 2 1 A,V,NMW rh72-node2 0 1 Qdevice [root@rh72-node1:~]# pcs quorum device status Qdevice information ------------------- Model: Net Node ID: 1 Configured node list: 0 Node ID = 1 1 Node ID = 2 Membership node list: 1, 2 Qdevice-net information ---------------------- Cluster name: rhel72 QNetd host: rh72-node3:5403 Algorithm: LMS Tie-breaker: Node with lowest node ID State: Connected > Check it from the quorum device side: [root@rh72-node3:~]# pcs qdevice status net --full QNetd address: *:5403 TLS: Supported (client certificate required) Connected clients: 2 Connected clusters: 1 Maximum send/receive size: 32768/32768 bytes Cluster "rhel72": Algorithm: LMS Tie-breaker: Node with lowest node ID Node ID 2: Client address: ::ffff:192.168.122.122:50028 HB interval: 8000ms Configured node list: 1, 2 Ring ID: 1.2050 Membership node list: 1, 2 TLS active: Yes (client certificate verified) Vote: ACK (ACK) Node ID 1: Client address: ::ffff:192.168.122.121:48786 HB interval: 8000ms Configured node list: 1, 2 Ring ID: 1.2050 Membership node list: 1, 2 TLS active: Yes (client certificate verified) Vote: ACK (ACK) > It is possible to change settings: [root@rh72-node1:~]# pcs quorum device update model algorithm=ffsplit Sending updated corosync.conf to nodes... rh72-node1: Succeeded rh72-node2: Succeeded Corosync configuration reloaded Reloading qdevice configuration on nodes... rh72-node1: corosync-qdevice stopped rh72-node2: corosync-qdevice stopped rh72-node1: corosync-qdevice started rh72-node2: corosync-qdevice started [root@rh72-node1:~]# pcs quorum device status --full Qdevice information ------------------- Model: Net Node ID: 1 HB interval: 10000ms Sync HB interval: 30000ms Configured node list: 0 Node ID = 1 1 Node ID = 2 Ring ID: 1.2050 Membership node list: 1, 2 Quorate: Yes Quorum node list: 0 Node ID = 1, State = member 1 Node ID = 2, State = member Expected votes: 3 Last poll call: 2016-06-29T13:22:15 (cast vote) Qdevice-net information ---------------------- Cluster name: rhel72 QNetd host: rh72-node3:5403 Connect timeout: 8000ms HB interval: 8000ms VQ vote timer interval: 5000ms TLS: Supported Algorithm: Fifty-Fifty split Tie-breaker: Node with lowest node ID Poll timer running: Yes (cast vote) State: Connected TLS active: Yes (client certificate sent) Connected since: 2016-06-29T13:20:30 Echo reply received: 2016-06-29T13:22:14 > When we do not want to use a quorum device any more: [root@rh72-node1:~]# pcs quorum device remove Sending updated corosync.conf to nodes... rh72-node1: Succeeded rh72-node2: Succeeded Corosync configuration reloaded Disabling corosync-qdevice... rh72-node1: corosync-qdevice disabled rh72-node2: corosync-qdevice disabled Stopping corosync-qdevice... rh72-node1: corosync-qdevice stopped rh72-node2: corosync-qdevice stopped Removing qdevice certificates from nodes... rh72-node1: Succeeded rh72-node2: Succeeded [root@rh72-node1:~]# pcs quorum device status Error: Unable to get quorum status: corosync-qdevice-tool: Can't connect to QDevice socket (is QDevice running?): No such file or directory [root@rh72-node1:~]# pcs quorum status Quorum information ------------------ Date: Wed Jun 29 13:23:03 2016 Quorum provider: corosync_votequorum Nodes: 2 Node ID: 1 Ring ID: 1/8272 Quorate: Yes Votequorum information ---------------------- Expected votes: 2 Highest expected: 2 Total votes: 2 Quorum: 1 Flags: 2Node Quorate Membership information ---------------------- Nodeid Votes Qdevice Name 1 1 NR rh72-node1 (local) 2 1 NR rh72-node2 [root@rh72-node1:~]# pcs quorum config Options: [root@rh72-node3:~]# pcs qdevice status net --full QNetd address: *:5403 TLS: Supported (client certificate required) Connected clients: 0 Connected clusters: 0 Maximum send/receive size: 32768/32768 bytes > We can destroy the quorum device: [root@rh72-node3:~]# pcs qdevice destroy net Stopping quorum device... quorum device stopped quorum device disabled Quorum device 'net' configuration files removed [root@rh72-node3:~]# pcs qdevice status net --full Error: Unable to get status of quorum device 'net': corosync-qnetd-tool: Can't connect to qnetd socket (is QNetd running?): No such file or directory > Apart from qdevice it is also possible to configure quorum options. This is only possible when corosync is not running: [root@rh72-node1:~]# pcs quorum config Options: [root@rh72-node1:~]# pcs quorum update wait_for_all=1 Checking corosync is not running on nodes... Error: rh72-node1: corosync is running Error: rh72-node2: corosync is running [root@rh72-node1:~]# pcs cluster stop --all rh72-node2: Stopping Cluster (pacemaker)... rh72-node1: Stopping Cluster (pacemaker)... rh72-node1: Stopping Cluster (corosync)... rh72-node2: Stopping Cluster (corosync)... [root@rh72-node1:~]# pcs quorum update wait_for_all=1 Checking corosync is not running on nodes... rh72-node2: corosync is not running rh72-node1: corosync is not running Sending updated corosync.conf to nodes... rh72-node1: Succeeded rh72-node2: Succeeded [root@rh72-node1:~]# pcs quorum config Options: wait_for_all: 1 qdevice is not supported by web UI yet Created attachment 1174429 [details]
proposed fix 11
built-in help and documentation fixes
additional patch - fixed tests https://github.com/ClusterLabs/pcs/commit/c178935290a5387fdfbd54330769ee159d8916b6 Any reason why "pcs quorum" is not interpreted as "pcs quorum config" (e.g., for consistency with "pcs alert")? On cluster-less machine: # pcs quorum device add model net host=rh72-node3 > Error: Unable to read /etc/corosync/corosync.conf: No such file or > directory would expect also something like: > You need to set cluster up first. On a machine with (dummy) corosync.conf but not with a running cluster: # pcs quorum device add model net host=localhost algorithm=lm > Setting up qdevice certificates on nodes... > Error: localhost: Unable to read certificate: > No such file or directory - /etc/corosync/qnetd/nssdb/qnetd-cacert.crt with --debug: Running: /usr/sbin/corosync -v > Finished running: /usr/sbin/corosync -v > Return value: 0 > --Debug Output Start-- > Corosync Cluster Engine, version '2.4.0' > Copyright (c) 2006-2009 Red Hat, Inc. > > --Debug Output End-- > > Running: /usr/bin/ruby -I/usr/lib/pcsd/ /usr/lib/pcsd/pcsd-cli.rb read_tokens > --Debug Input Start-- > {} > --Debug Input End-- > Return Value: 0 > --Debug Output Start-- > { > "status": "ok", > "data": { > "localhost": "413253e2-2cf3-4c0f-a631-5228ca2dbf22" > }, > "log": [ > "I, [2016-07-11T12:41:55.120966 #17607] INFO -- : PCSD Debugging enabled\n", > "D, [2016-07-11T12:41:55.121088 #17607] DEBUG -- : Did not detect RHEL 6\n", > "I, [2016-07-11T12:41:55.121127 #17607] INFO -- : Running: /usr/sbin/corosync-cmapctl totem.cluster_n > ame\n", > "I, [2016-07-11T12:41:55.121155 #17607] INFO -- : CIB USER: hacluster, groups: \n", > "D, [2016-07-11T12:41:55.126300 #17607] DEBUG -- : []\n", > "D, [2016-07-11T12:41:55.126382 #17607] DEBUG -- : [\"Failed to initialize the cmap API. Error CS_ERR_ > LIBRARY\\n\"]\n", > "D, [2016-07-11T12:41:55.126449 #17607] DEBUG -- : Duration: 0.00510652s\n", > "I, [2016-07-11T12:41:55.126526 #17607] INFO -- : Return Value: 1\n" > ] > } > --Debug Output End-- > > Setting up qdevice certificates on nodes... > Sending HTTP Request to: https://localhost:2224/remote/qdevice_net_get_ca_certificate > > Finished calling: https://localhost:2224/remote/qdevice_net_get_ca_certificate > Response Code: 400 > --Debug Response Start-- > Unable to read certificate: No such file or directory - /etc/corosync/qnetd/nssdb/qnetd-cacert.crt > --Debug Response End-- > > Error: localhost: Unable to read certificate: No such file or directory - /etc/corosync/qnetd/nssdb/qnetd- > cacert.crt [comment 25] got clarified, corosync not running will only cause "read_tokens" to be verbose about that, but that's irrelevant (contrary to my expectations) to the final message, which is actually the root cause (even a bit cryptic to the novice user). Created attachment 1178400 [details]
proposed fix 12
Added "pcs status quorum" and "pcs status qdevice" commands. Command "pcs quorum" now displays quorum configuration instead of pcs help.
(In reply to Jan Pokorný from comment #24) > On a machine with (dummy) corosync.conf but not with a running cluster: > > # pcs quorum device add model net host=localhost algorithm=lm > > Setting up qdevice certificates on nodes... > > Error: localhost: Unable to read certificate: > > No such file or directory - /etc/corosync/qnetd/nssdb/qnetd-cacert.crt > This is expected. Quorum device needs to be created first by "pcs qdevice setup" command. This is mentioned in pcs help. If only modification of a standalone corosync.conf is desired, use --corosync_conf=<path to corosync.conf> option. # pcs quorum -h | sed -n '/device add/{:s;s/ //;p;n;/^$/d;bs}'
> device add [<generic options>] model <device model> [<model options>]
> Add a quorum device to the cluster. Quorum device needs to be created
> first by "pcs qdevice setup" command. It is not possible to use more
> than one quorum device in a cluster simultaneously. Generic options,
> model and model options are all documented in corosync's
> corosync-qdevice(8) man page.
I don't follow why it stresses "corosync" out (already apparent from the
man page title). It may also be a bit misleading as for RHEL 7, the
man page is in corosync-qdevice package.
For the release note description of this feature (in the Doc Text field) I substituted the introductory paragraph from the documentation for this feature (which has already been through review in that context). I think this introduces the features and its use. Created attachment 1199455 [details]
proposed fix 13
Created attachment 1199456 [details]
proposed fix 14
*** Bug 1368106 has been marked as a duplicate of this bug. *** Created attachment 1201629 [details]
proposed fix 15
Fixed following issues:
1) pcs exits with an error when qdevice is already stopped:
[root@virt-293 ~]# pcs qdevice setup model net
Quorum device 'net' initialized
[root@virt-293 ~]# pcs qdevice destroy net
Error: Unable to get status of quorum device 'net': corosync-qnetd-tool: Can't connect to qnetd socket (is QNetd running?): No such file or directory
[root@virt-293 ~]# pcs qdevice stop net
Error: Unable to get status of quorum device 'net': corosync-qnetd-tool: Can't connect to qnetd socket (is QNetd running?): No such file or directory
[root@virt-293 ~]# echo $?
1
2) corosync-qdevice gets restarted in "pcs quorum device remove":
[root@virt-294 ~]# pcs quorum device remove
Sending updated corosync.conf to nodes...
virt-294: Succeeded
virt-296: Succeeded
virt-295: Succeeded
Corosync configuration reloaded
Disabling corosync-qdevice...
virt-294: corosync-qdevice disabled
virt-296: corosync-qdevice disabled
virt-295: corosync-qdevice disabled
Stopping corosync-qdevice...
virt-295: corosync-qdevice stopped
virt-294: corosync-qdevice stopped
virt-296: corosync-qdevice stopped
Removing qdevice certificates from nodes...
virt-296: Succeeded
virt-294: Succeeded
virt-295: Succeeded
[root@virt-294 ~]# systemctl status corosync-qdevice
● corosync-qdevice.service - Corosync Qdevice daemon
Loaded: loaded (/usr/lib/systemd/system/corosync-qdevice.service; disabled; vendor preset: disabled)
Active: inactive (dead)
Sep 16 13:13:52 virt-294.cluster-qe.lab.eng.brq.redhat.com systemd[1]: Stopping Corosync Qdevice daemon...
Sep 16 13:13:52 virt-294.cluster-qe.lab.eng.brq.redhat.com corosync-qdevice[25963]: Signaling Corosync Qdevice daemon (corosync-qdevice) to terminate: [ OK ]
Sep 16 13:13:52 virt-294.cluster-qe.lab.eng.brq.redhat.com corosync-qdevice[25963]: Waiting for corosync-qdevice services to unload:[ OK ]
Sep 16 13:13:52 virt-294.cluster-qe.lab.eng.brq.redhat.com systemd[1]: Stopped Corosync Qdevice daemon.
Sep 16 13:13:53 virt-294.cluster-qe.lab.eng.brq.redhat.com systemd[1]: Starting Corosync Qdevice daemon...
Sep 16 13:13:53 virt-294.cluster-qe.lab.eng.brq.redhat.com corosync-qdevice[25995]: Starting Corosync Qdevice daemon (corosync-qdevice): [ OK ]
Sep 16 13:13:53 virt-294.cluster-qe.lab.eng.brq.redhat.com systemd[1]: Started Corosync Qdevice daemon.
Sep 16 13:14:12 virt-294.cluster-qe.lab.eng.brq.redhat.com corosync-qdevice[26001]: 50:50 split algorithm works only if qdevice has exactly 1 vote. Node 1 has 0 votes!
Sep 16 13:14:12 virt-294.cluster-qe.lab.eng.brq.redhat.com corosync-qdevice[26001]: 50:50 split algorithm works only if qdevice has exactly 1 vote. Node 1 has 0 votes!
Sep 16 13:14:12 virt-294.cluster-qe.lab.eng.brq.redhat.com corosync-qdevice[26001]: 50:50 split algorithm works only if qdevice has exactly 1 vote. Node 1 has 0 votes!
[root@virt-294 ~]# systemctl is-active corosync-qdevice
Before Fix: [vm-rhel72-1 ~] $ rpm -q pcs pcs-0.9.152-9.el7.x86_64 [vm-rhel72-1 ~] $ pcs qdevice setup model net Quorum device 'net' initialized [vm-rhel72-1 ~] $ pcs qdevice destroy net Error: Unable to get status of quorum device 'net': corosync-qnetd-tool: Can't connect to qnetd socket (is QNetd running?): No such file or directory [vm-rhel72-1 ~] $ echo $? 1 [vm-rhel72-1 ~] $ pcs qdevice stop net Error: Unable to get status of quorum device 'net': corosync-qnetd-tool: Can't connect to qnetd socket (is QNetd running?): No such file or directory [vm-rhel72-1 ~] $ echo $? 1 After Fix: [vm-rhel72-1 ~] $ rpm -q pcs pcs-0.9.152-10.el7.x86_64 [vm-rhel72-1 ~] $ pcs qdevice setup model net Quorum device 'net' initialized [vm-rhel72-1 ~] $ pcs qdevice stop net Stopping quorum device... quorum device stopped [vm-rhel72-1 ~] $ pcs qdevice destroy net Stopping quorum device... quorum device stopped quorum device disabled Quorum device 'net' configuration files removed Adding "QDevice" to the release note description since users might be searching on that particular term (event though that's not the term currently used in the doc). Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://rhn.redhat.com/errata/RHSA-2016-2596.html Updating doc text to indicate that this feature is being removed from Technology Preview for RHEL 7.4 and is now fully supported. Explicitly noting in doc text that this feature was previously available as Technology Preview. |