Re: [ClusterLabs] Stupid DRBD/LVM Global Filter Question
Roger -- Thank you, sir. That does help. -Original Message- From: Roger Zhou Sent: Wednesday, October 30, 2019 2:56 AM To: Cluster Labs - All topics related to open-source clustering welcomed ; Eric Robinson Subject: Re: [ClusterLabs] Stupid DRBD/LVM Global Filter Question On 10/30/19 6:17 AM, Eric Robinson wrote: > If I have an LV as a backing device for a DRBD disk, can someone > explain why I need an LVM filter? It seems to me that we would want > the LV to be always active under both the primary and secondary DRBD > devices, and there should be no need or desire to have the LV > activated or deactivated by Pacemaker. What am I missing? Your understanding is correct. No need to use LVM resource agent from Pacemaker in your case. --Roger > > --Eric > > Disclaimer : This email and any files transmitted with it are > confidential and intended solely for intended recipients. If you are > not the named addressee you should not disseminate, distribute, copy > or alter this email. Any views or opinions presented in this email are > solely those of the author and might not represent those of Physician > Select Management. Warning: Although Physician Select Management has > taken reasonable precautions to ensure no viruses are present in this > email, the company cannot accept responsibility for any loss or damage > arising from the use of this email or attachments. > > ___ > Manage your subscription: > https://lists.clusterlabs.org/mailman/listinfo/users > > ClusterLabs home: https://www.clusterlabs.org/ > Disclaimer : This email and any files transmitted with it are confidential and intended solely for intended recipients. If you are not the named addressee you should not disseminate, distribute, copy or alter this email. Any views or opinions presented in this email are solely those of the author and might not represent those of Physician Select Management. Warning: Although Physician Select Management has taken reasonable precautions to ensure no viruses are present in this email, the company cannot accept responsibility for any loss or damage arising from the use of this email or attachments. ___ Manage your subscription: https://lists.clusterlabs.org/mailman/listinfo/users ClusterLabs home: https://www.clusterlabs.org/
Re: [ClusterLabs] fencing on iscsi device not working
30.10.2019 15:46, RAM PRASAD TWISTED ILLUSIONS пишет: > Hi everyone, > > I am trying to set up a storage cluster with two nodes, both running debian > buster. The two nodes called, duke and miles, have a LUN residing on a SAN > box as their shared storage device between them. As you can see in the > output of pcs status, all the demons are active and I can get the nodes > online without any issues. However, I cannot get the fencing resources to > start. > > These two nodes were running debian jessie before and had access to the > same LUN in a storage cluster configuration. Now, I am trying to recreate a > similar setup with both nodes now running the latest debian. I am not sure > if this is relevant, but this LUN already has shared VG with data on it. I > am wondering if this could be the cause of the trouble? Should I be > creating my stonith device on a different/fresh LUN? > > ### pcs status > Cluster name: jazz > Stack: corosync > Current DC: duke (version 2.0.1-9e909a5bdd) - partition with quorum > Last updated: Wed Oct 30 11:58:19 2019 > Last change: Wed Oct 30 11:28:28 2019 by root via cibadmin on duke > > 2 nodes configured > 2 resources configured > > Online: [ duke miles ] > > Full list of resources: > > fence_duke(stonith:fence_scsi):Stopped > fence_miles(stonith:fence_scsi):Stopped > > Failed Fencing Actions: > * unfencing of duke failed: delegate=, client=pacemaker-controld.1703, > origin=duke, > last-failed='Wed Oct 30 11:43:29 2019' > * unfencing of miles failed: delegate=, client=pacemaker-controld.1703, > origin=duke, > last-failed='Wed Oct 30 11:43:29 2019' > > Daemon Status: > corosync: active/enabled > pacemaker: active/enabled > pcsd: active/enabled > ### > > I used the following commands to add the two fencing devices and set their > location constraints . > > ### > sudo pcs cluster cib test_cib_cfg > pcs -f test_cib_cfg stonith create fence_duke fence_scsi > pcmk_host_list=duke pcmk_reboot_action="off" According to documentation, pcmk_host_list is used only if pcmk_host_check=static-list which is not default, by default pacemaker queries agent for nodes it can fence and fence_scsi does not return anything. ___ Manage your subscription: https://lists.clusterlabs.org/mailman/listinfo/users ClusterLabs home: https://www.clusterlabs.org/
Re: [ClusterLabs] fencing on iscsi device not working
On Wed, 2019-10-30 at 13:46 +0100, RAM PRASAD TWISTED ILLUSIONS wrote: > Hi everyone, > > I am trying to set up a storage cluster with two nodes, both running > debian buster. The two nodes called, duke and miles, have a LUN > residing on a SAN box as their shared storage device between them. As > you can see in the output of pcs status, all the demons are active > and I can get the nodes online without any issues. However, I cannot > get the fencing resources to start. > > These two nodes were running debian jessie before and had access to > the same LUN in a storage cluster configuration. Now, I am trying to > recreate a similar setup with both nodes now running the latest > debian. I am not sure if this is relevant, but this LUN already has > shared VG with data on it. I am wondering if this could be the cause > of the trouble? Should I be creating my stonith device on a > different/fresh LUN? > > ### pcs status > Cluster name: jazz > Stack: corosync > Current DC: duke (version 2.0.1-9e909a5bdd) - partition with quorum > Last updated: Wed Oct 30 11:58:19 2019 > Last change: Wed Oct 30 11:28:28 2019 by root via cibadmin on duke > > 2 nodes configured > 2 resources configured > > Online: [ duke miles ] > > Full list of resources: > > fence_duke(stonith:fence_scsi):Stopped > fence_miles(stonith:fence_scsi):Stopped > > Failed Fencing Actions: > * unfencing of duke failed: delegate=, client=pacemaker- > controld.1703, origin=duke, > last-failed='Wed Oct 30 11:43:29 2019' > * unfencing of miles failed: delegate=, client=pacemaker- > controld.1703, origin=duke, > last-failed='Wed Oct 30 11:43:29 2019' > > Daemon Status: > corosync: active/enabled > pacemaker: active/enabled > pcsd: active/enabled > ### > > I used the following commands to add the two fencing devices and set > their location constraints . > > ### > sudo pcs cluster cib test_cib_cfg > pcs -f test_cib_cfg stonith create fence_duke fence_scsi > pcmk_host_list=duke pcmk_reboot_action="off" devices="/dev/disk/by- > id/wwn-0x600c0ff0001e8e3c89601b580100" meta provides="unfencing" > pcs -f test_cib_cfg stonith create fence_miles fence_scsi > pcmk_host_list=miles pcmk_reboot_action="off" devices="/dev/disk/by- > id/wwn-0x600c0ff0001e8e3c89601b580100" delay=15 meta > provides="unfencing" > pcs -f test_cib_cfg constraint location fence_duke avoids > duke=INFINITY > pcs -f test_cib_cfg constraint location fence_miles avoids > miles=INFINITY Use a score less than INFINITY. You want the devices to be able to run on the target node if for some reason the other node is unable to (e.g. if it's in standby). I'm not sure whether that will fix the issue here (see below). > pcs cluster cib-push test_cib_cfg > ### > > Here is the output in /var/log/pacemaker/pacemaker.log after adding > the fencing resources > > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] > (determine_online_status_fencing) info: Node miles is active > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] > (determine_online_status) info: Node miles is online > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] > (determine_online_status_fencing) info: Node duke is active > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] > (determine_online_status) info: Node duke is online > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] > (unpack_node_loop) info: Node 2 is already processed > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] > (unpack_node_loop) info: Node 1 is already processed > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] > (unpack_node_loop) info: Node 2 is already processed > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] > (unpack_node_loop) info: Node 1 is already processed > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (common_print) info: > fence_duke(stonith:fence_scsi): Stopped > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (common_print) info: > fence_miles (stonith:fence_scsi): Stopped > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (RecurringOp) info: > Start recurring monitor (60s) for fence_duke on miles > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (RecurringOp) info: > Start recurring monitor (60s) for fence_miles on duke > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] > (LogNodeActions)notice: * Fence (on) miles 'required by > fence_duke monitor' > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] > (LogNodeActions)notice: * Fence (on) duke 'required by > fence_duke monitor' > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (LogAction) notice: > * Start fence_duke ( miles ) > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (LogAction) notice: > * Start fence_miles( duke ) > Oct 30 12:06:02 duke pacemaker-schedulerd[1702] > (process_pe_message)notice: Calculated transition 63, saving > inputs in /var/lib/pacemaker/pengine/pe-input-23.bz2 > Oc
[ClusterLabs] fencing on iscsi device not working
Hi everyone, I am trying to set up a storage cluster with two nodes, both running debian buster. The two nodes called, duke and miles, have a LUN residing on a SAN box as their shared storage device between them. As you can see in the output of pcs status, all the demons are active and I can get the nodes online without any issues. However, I cannot get the fencing resources to start. These two nodes were running debian jessie before and had access to the same LUN in a storage cluster configuration. Now, I am trying to recreate a similar setup with both nodes now running the latest debian. ### pcs status Cluster name: jazz Stack: corosync Current DC: duke (version 2.0.1-9e909a5bdd) - partition with quorum Last updated: Wed Oct 30 11:58:19 2019 Last change: Wed Oct 30 11:28:28 2019 by root via cibadmin on duke 2 nodes configured 2 resources configured Online: [ duke miles ] Full list of resources: fence_duke (stonith:fence_scsi): Stopped fence_miles (stonith:fence_scsi): Stopped Failed Fencing Actions: * unfencing of duke failed: delegate=, client=pacemaker-controld.1703, origin=duke, last-failed='Wed Oct 30 11:43:29 2019' * unfencing of miles failed: delegate=, client=pacemaker-controld.1703, origin=duke, last-failed='Wed Oct 30 11:43:29 2019' Daemon Status: corosync: active/enabled pacemaker: active/enabled pcsd: active/enabled ### I used the following commands to add the two fencing devices and set their location constraints . ### sudo pcs cluster cib test_cib_cfg pcs -f test_cib_cfg stonith create fence_duke fence_scsi pcmk_host_list=duke pcmk_reboot_action="off" devices="/dev/disk/by-id/wwn-0x600c0ff0001e8e3c89601b580100" meta provides="unfencing" pcs -f test_cib_cfg stonith create fence_miles fence_scsi pcmk_host_list=miles pcmk_reboot_action="off" devices="/dev/disk/by-id/wwn-0x600c0ff0001e8e3c89601b580100" delay=15 meta provides="unfencing" pcs -f test_cib_cfg constraint location fence_duke avoids duke=INFINITY pcs -f test_cib_cfg constraint location fence_miles avoids miles=INFINITY pcs cluster cib-push test_cib_cfg ### Here is the output in /var/log/pacemaker/pacemaker.log after adding the fencing resources Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (determine_online_status_fencing) info: Node miles is active Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (determine_online_status) info: Node miles is online Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (determine_online_status_fencing) info: Node duke is active Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (determine_online_status) info: Node duke is online Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (unpack_node_loop) info: Node 2 is already processed Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (unpack_node_loop) info: Node 1 is already processed Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (unpack_node_loop) info: Node 2 is already processed Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (unpack_node_loop) info: Node 1 is already processed Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (common_print) info: fence_duke (stonith:fence_scsi): Stopped Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (common_print) info: fence_miles (stonith:fence_scsi): Stopped Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (RecurringOp) info: Start recurring monitor (60s) for fence_duke on miles Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (RecurringOp) info: Start recurring monitor (60s) for fence_miles on duke Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (LogNodeActions) notice: * Fence (on) miles 'required by fence_duke monitor' Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (LogNodeActions) notice: * Fence (on) duke 'required by fence_duke monitor' Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (LogAction) notice: * Start fence_duke ( miles ) Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (LogAction) notice: * Start fence_miles ( duke ) Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (process_pe_message) notice: Calculated transition 63, saving inputs in /var/lib/pacemaker/pengine/pe-input-23.bz2 Oct 30 12:06:02 duke pacemaker-controld [1703] (do_state_transition) info: State transition S_POLICY_ENGINE -> S_TRANSITION_ENGINE | input=I_PE_SUCCESS cause=C_IPC_MESSAGE origin=handle_response Oct 30 12:06:02 duke pacemaker-controld [1703] (do_te_invoke) info: Processing graph 63 (ref=pe_calc-dc-1572433562-101) derived from /var/lib/pacemaker/pengine/pe-input-23.bz2 Oct 30 12:06:02 duke pacemaker-controld [1703] (te_fence_node) notice: Requesting fencing (on) of node miles | action=5 timeout=6 Oct 30 12:06:02 duke pacemaker-controld [1703] (te_fence_node) notice: Requesting fencing (on) of node duke | action=2 timeout=6 Oct 30 12:06:02 duke pacemaker-fenced [1699] (handle_request) no
[ClusterLabs] Antw: Stupid DRBD/LVM Global Filter Question
>>> Eric Robinson schrieb am 29.10.2019 um 23:17 in Nachricht > If I have an LV as a backing device for a DRBD disk, can someone explain why > I need an LVM filter? It seems to me that we would want the LV to be always > active under both the primary and secondary DRBD devices, and there should be > no need or desire to have the LV activated or deactivated by Pacemaker. What > am I missing? At least in the past it also had a performance reason: Most LVM tools did not cache device information, and instead did O_DIRECT device access. In a SAN environment with more than 100 disk paths, plus device-mapper devices, partitions, etc. The delay and I/O impact was significant. So if you are using LVM only on a few devices, help LVM to locate those. Also for multipath you want LVM to use the multipath device instead of a single path device for example. Regards, Ulrich > > ‑‑Eric > > > > > Disclaimer : This email and any files transmitted with it are confidential > and intended solely for intended recipients. If you are not the named > addressee you should not disseminate, distribute, copy or alter this email. > Any views or opinions presented in this email are solely those of the author > and might not represent those of Physician Select Management. Warning: > Although Physician Select Management has taken reasonable precautions to > ensure no viruses are present in this email, the company cannot accept > responsibility for any loss or damage arising from the use of this email or > attachments. ___ Manage your subscription: https://lists.clusterlabs.org/mailman/listinfo/users ClusterLabs home: https://www.clusterlabs.org/
[ClusterLabs] fencing on iscsi device not working
Hi everyone, I am trying to set up a storage cluster with two nodes, both running debian buster. The two nodes called, duke and miles, have a LUN residing on a SAN box as their shared storage device between them. As you can see in the output of pcs status, all the demons are active and I can get the nodes online without any issues. However, I cannot get the fencing resources to start. These two nodes were running debian jessie before and had access to the same LUN in a storage cluster configuration. Now, I am trying to recreate a similar setup with both nodes now running the latest debian. I am not sure if this is relevant, but this LUN already has shared VG with data on it. I am wondering if this could be the cause of the trouble? Should I be creating my stonith device on a different/fresh LUN? ### pcs status Cluster name: jazz Stack: corosync Current DC: duke (version 2.0.1-9e909a5bdd) - partition with quorum Last updated: Wed Oct 30 11:58:19 2019 Last change: Wed Oct 30 11:28:28 2019 by root via cibadmin on duke 2 nodes configured 2 resources configured Online: [ duke miles ] Full list of resources: fence_duke(stonith:fence_scsi):Stopped fence_miles(stonith:fence_scsi):Stopped Failed Fencing Actions: * unfencing of duke failed: delegate=, client=pacemaker-controld.1703, origin=duke, last-failed='Wed Oct 30 11:43:29 2019' * unfencing of miles failed: delegate=, client=pacemaker-controld.1703, origin=duke, last-failed='Wed Oct 30 11:43:29 2019' Daemon Status: corosync: active/enabled pacemaker: active/enabled pcsd: active/enabled ### I used the following commands to add the two fencing devices and set their location constraints . ### sudo pcs cluster cib test_cib_cfg pcs -f test_cib_cfg stonith create fence_duke fence_scsi pcmk_host_list=duke pcmk_reboot_action="off" devices="/dev/disk/by-id/wwn-0x600c0ff0001e8e3c89601b580100" meta provides="unfencing" pcs -f test_cib_cfg stonith create fence_miles fence_scsi pcmk_host_list=miles pcmk_reboot_action="off" devices="/dev/disk/by-id/wwn-0x600c0ff0001e8e3c89601b580100" delay=15 meta provides="unfencing" pcs -f test_cib_cfg constraint location fence_duke avoids duke=INFINITY pcs -f test_cib_cfg constraint location fence_miles avoids miles=INFINITY pcs cluster cib-push test_cib_cfg ### Here is the output in /var/log/pacemaker/pacemaker.log after adding the fencing resources Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (determine_online_status_fencing) info: Node miles is active Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (determine_online_status) info: Node miles is online Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (determine_online_status_fencing) info: Node duke is active Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (determine_online_status) info: Node duke is online Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (unpack_node_loop) info: Node 2 is already processed Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (unpack_node_loop) info: Node 1 is already processed Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (unpack_node_loop) info: Node 2 is already processed Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (unpack_node_loop) info: Node 1 is already processed Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (common_print) info: fence_duke(stonith:fence_scsi): Stopped Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (common_print) info: fence_miles (stonith:fence_scsi): Stopped Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (RecurringOp) info: Start recurring monitor (60s) for fence_duke on miles Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (RecurringOp) info: Start recurring monitor (60s) for fence_miles on duke Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (LogNodeActions) notice: * Fence (on) miles 'required by fence_duke monitor' Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (LogNodeActions) notice: * Fence (on) duke 'required by fence_duke monitor' Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (LogAction) notice: * Start fence_duke ( miles ) Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (LogAction) notice: * Start fence_miles( duke ) Oct 30 12:06:02 duke pacemaker-schedulerd[1702] (process_pe_message) notice: Calculated transition 63, saving inputs in /var/lib/pacemaker/pengine/pe-input-23.bz2 Oct 30 12:06:02 duke pacemaker-controld [1703] (do_state_transition) info: State transition S_POLICY_ENGINE -> S_TRANSITION_ENGINE | input=I_PE_SUCCESS cause=C_IPC_MESSAGE origin=handle_response Oct 30 12:06:02 duke pacemaker-controld [1703] (do_te_invoke) info: Processing graph 63 (ref=pe_calc-dc-1572433562-101) derived from /var/lib/pacemaker/pengine/pe-input-23.bz2 Oct 30 12:06:02 duke pacemaker-controld [1703] (te_fence_node) notice: Requesting fencing (on) of node miles | action=5 timeout=6 Oct 30 12:06:02 duke pacemaker-controld [1703] (te_fence_node) notice: Requesting fencing (on) of node
Re: [ClusterLabs] Stupid DRBD/LVM Global Filter Question
On 10/30/19 6:17 AM, Eric Robinson wrote: > If I have an LV as a backing device for a DRBD disk, can someone explain > why I need an LVM filter? It seems to me that we would want the LV to be > always active under both the primary and secondary DRBD devices, and > there should be no need or desire to have the LV activated or > deactivated by Pacemaker. What am I missing? Your understanding is correct. No need to use LVM resource agent from Pacemaker in your case. --Roger > > --Eric > > Disclaimer : This email and any files transmitted with it are > confidential and intended solely for intended recipients. If you are not > the named addressee you should not disseminate, distribute, copy or > alter this email. Any views or opinions presented in this email are > solely those of the author and might not represent those of Physician > Select Management. Warning: Although Physician Select Management has > taken reasonable precautions to ensure no viruses are present in this > email, the company cannot accept responsibility for any loss or damage > arising from the use of this email or attachments. > > ___ > Manage your subscription: > https://lists.clusterlabs.org/mailman/listinfo/users > > ClusterLabs home: https://www.clusterlabs.org/ > ___ Manage your subscription: https://lists.clusterlabs.org/mailman/listinfo/users ClusterLabs home: https://www.clusterlabs.org/