Forum
Welcome, Guest
Username: Password: Remember me
This is the optional category header for the Suggestion Box.
  • Page:
  • 1

TOPIC:

Installed new NIC, now Slave in WFConnection State 6 years 1 month ago #1558

  • Check Signal
  • Check Signal's Avatar Topic Author
  • Offline
  • Posts: 3
We were having consistent issues with the network connectivity in one of our hosts so we took it down and replaced the NIC, now the primary hypervisor says:
--------------------------------------------------------
| iSCSI-HA Status: Running 4418                        |
| Last Updated: Fri Feb 16 16:23:34 CST 2018           |
| HOST ROLE:              SLAVE                        |
| VIRTUAL IP:             10.10.10.3 is not local      |
| ISCSI TARGET:           Stopped [expected stopped]   |
| DRBD ROLE:              iscsi1=Secondary             |
| DRBD CONNECTION:        iscsi1 in WFConnection state |
--------------------------------------------------------
Control + C to exit


---------------
| DRBD Status |
---------------

-------------------------------------------------
| version: 8.4.5 (api:1/proto:86-101)                                         |
| srcversion: 2A6B2FA4F0703B49CA9C727                                         |
|  1: cs:WFConnection ro:Secondary/Unknown ds:UpToDate/DUnknown C r-----      |
|     ns:0 nr:0 dw:0 dr:0 al:0 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:f oos:5066752 |
--------------------------------------------------

We tried pinging 10.10.10.3 on this host and it responded.

I noticed Salvatore mentioned we should run this on the primary host to get them to resync:

drbdadm -- --overwrite-data-of-peer primary iscsi1

But I'm afraid to run that considering it's saying 10.10.10.3 isn't local.

Please Log in or Create an account to join the conversation.

Installed new NIC, now Slave in WFConnection State 6 years 1 month ago #1559

  • Salvatore Costantino
  • Salvatore Costantino's Avatar
  • Offline
  • Posts: 722
It appears that your primary server has been demoted to the slave role and the floating ip is live on the other host

Chances are the other host had issues starting DRBD. Try shutting down services tnd then start again with the following commands:

service iscsi-ha-watchdog
service iscsi-ha stop
service drbd stop
systemctl forcedstop tgtd
service iscsi-ha-watchdog start

Also, your situation does not require a resync like the other post you referred to. In the other case, there was no data on a newly introduced host. Your situation is different in that your data is intact on both hosts and will be brought up to data as soon as drbd connects.

Please Log in or Create an account to join the conversation.

Installed new NIC, now Slave in WFConnection State 6 years 1 month ago #1560

  • Check Signal
  • Check Signal's Avatar Topic Author
  • Offline
  • Posts: 3
Thank you for your response, Salvatore! Unfortunately, that didn't do it. I ran the commands you posted on just the hypervisor that I'm talking about (one command failed, systemctl forcedstop tgtd didn't really match anything that I tabbed into).

It brought me back to this. Should I run it on both hypervisors? I should also mention that this hypervisor doesn't have any of the hosts running on it right now. They're all running on the other server.
--------------------------------
| iSCSI-HA Version IHA_2.1.5   |
| Sun Feb 18 16:57:55 CST 2018 |
--------------------------------
--------------------------------------------------------
| iSCSI-HA Status: Running 19221                       |
| Last Updated: Sun Feb 18 16:57:50 CST 2018           |
| HOST ROLE:              SLAVE                        |
| VIRTUAL IP:             10.10.10.3 is not local      |
| ISCSI TARGET:           Stopped [expected stopped]   |
| DRBD ROLE:              iscsi1=Secondary             |
| DRBD CONNECTION:        iscsi1 in WFConnection state |
--------------------------------------------------------
Control + C to exit


---------------
| DRBD Status |
---------------
-------------------------------------------------------------------------------
| version: 8.4.5 (api:1/proto:86-101)                                         |
| srcversion: 2A6B2FA4F0703B49CA9C727                                         |
|  1: cs:WFConnection ro:Secondary/Unknown ds:UpToDate/DUnknown C r-----      |

Please Log in or Create an account to join the conversation.

Last edit: by Check Signal.

Installed new NIC, now Slave in WFConnection State 6 years 1 month ago #1562

  • Salvatore Costantino
  • Salvatore Costantino's Avatar
  • Offline
  • Posts: 722
If the floating IP is pingable from both hosts, you can try running the following repair script. The script will prompt you to answer a few questions. Assuming that the current master has the most up to date data, select the master as being the survivor when prompted. Just in case you run it in the wrong direction, I would encourage you to export your vms first.

cd /etc/iscsi-ha/scripts
./drbd-sb-tool
The following user(s) said Thank You: Jiri Ledvinka

Please Log in or Create an account to join the conversation.

Installed new NIC, now Slave in WFConnection State 6 years 3 weeks ago #1573

  • Check Signal
  • Check Signal's Avatar Topic Author
  • Offline
  • Posts: 3
Brilliant! That did it. Now to still diagnose the real issue. Thank you so much for your help, Salvatore.

Please Log in or Create an account to join the conversation.

  • Page:
  • 1