LUN presented over FC not visible after ESXi Reboot

Status
Not open for further replies.

andstein85

Dabbler
Joined
Apr 21, 2017
Messages
14
Hello all!

I've for some time now had a 10TB zvol served as an RDM via a Qlogic QLE2462 FC card to a VM running on an ESXi 6.0u2 host.
FreeNAS Server Specs:
SuperMicro X7DWE board
2x Intel Xeon E5335
4x4GB FBDIMM DDR2
2xIBM M1015 SAS/SATA cards
6x 3TB 7200rpm white label disks

The ESXi host is an IBM x3650 m2

Everything has worked really well and performance is pretty great, however I have a pretty annoying problem with it and was wondering if anyone knew of a fix.

Basically everytime I reboot the ESXi host, when it comes back up, the LUN is no longer presented to the host and in order to fix it, I have to run the following cmds on the FreeNAS server:

[root@freenas] /# ctladm port -o off -t fc
Front End Ports disabled
[root@freenas] /# ctladm port -o on -t fc
Front End Ports enabled


This works no problem, but becomes very cumbersome when trying to do this from a smartphone...

Here is the pertinent dmesg logs that show the FC link going down:
Code:
isp0: Chan 0 LINK FAILED
isp1: Chan 0 LINK FAILED
ctlfeasync: WWPN 0x21000024ff467b9a port 0x0000ef path 1 target 0 left
ctlfeasync: WWPN 0x21000024ff467b9b port 0x0000ef path 0 target 0 left


And the subsequent log showing the host's FC link coming back up:
Code:
ctlfeasync: WWPN 0x21000024ff467b9a port 0x0000ef path 1 target 0 arrived
ctlfeasync: WWPN 0x21000024ff467b9b port 0x0000ef path 0 target 0 arrived


So even though FreeNAS see's the FC link from the ESXi host come online, the FC target still has to be reinitialized manually.

Anyone have any ideas on a better fix than what I've already come up with?
 
D

dlavigne

Guest
Have you tried adding that command to Tasks → Init/Shutdown Scripts?
 

andstein85

Dabbler
Joined
Apr 21, 2017
Messages
14
Yes, I have "ctladm port -o on -t fc" set as an init/shutdown script already.

That doesn't apply to this situation though because this occurs after rebooting an ESXi host that is served storage from FreeNAS, not after rebooting the FreeNAS box. Actually, rebooting the FreeNAS box "re-presents" the LUN in question in ESXi. But again, I'd rather not have to reboot my FreeNAS box every time I need to reboot my ESXi host, just like I'd rather not have to run the 2 cmd's referenced in my OP every time I reboot my ESXi host...
 

mav@

iXsystems
iXsystems
Joined
Sep 29, 2011
Messages
1,428
This behavior looks odd to me. I don't remember seeing such one on my tests. I can only say that FC code is constantly improving, so you may retest that on FreeNAS 11 or its already present nightly builds.
 

andstein85

Dabbler
Joined
Apr 21, 2017
Messages
14
I've upgraded my host to ESXi 6.5 and my storage box to FreeNAS 11, still persists... Its not ESXi, I know that for sure.

root@freenas:/ # sysctl dev.isp
dev.isp.1.topo: 2
dev.isp.1.loopstate: 10
dev.isp.1.fwstate: 3
dev.isp.1.linkstate: 1
dev.isp.1.speed: 4
dev.isp.1.role: 1
dev.isp.1.gone_device_time: 30
dev.isp.1.loop_down_limit: 60

dev.isp.1.wwpn: 2377900762153240051
dev.isp.1.wwnn: 2305843168115312115
dev.isp.1.%parent: pci3
dev.isp.1.%pnpinfo: vendor=0x1077 device=0x2432 subvendor=0x1077 subdevice=0x0138 class=0x0c0400
dev.isp.1.%location: slot=0 function=1 dbsf=pci0:3:0:1
dev.isp.1.%driver: isp
dev.isp.1.%desc: Qlogic ISP 2432 PCI FC-AL Adapter
dev.isp.0.topo: 2
dev.isp.0.loopstate: 10
dev.isp.0.fwstate: 3
dev.isp.0.linkstate: 1
dev.isp.0.speed: 4
dev.isp.0.role: 1
dev.isp.0.gone_device_time: 30
dev.isp.0.loop_down_limit: 60

dev.isp.0.wwpn: 2377900762153240050
dev.isp.0.wwnn: 2305843168115312114
dev.isp.0.%parent: pci3
dev.isp.0.%pnpinfo: vendor=0x1077 device=0x2432 subvendor=0x1077 subdevice=0x0138 class=0x0c0400
dev.isp.0.%location: slot=0 function=0 dbsf=pci0:3:0:0
dev.isp.0.%driver: isp
dev.isp.0.%desc: Qlogic ISP 2432 PCI FC-AL Adapter
dev.isp.%parent:


Anyone know if those 2 settings could have something to do with this problem? Maybe its a setting on the actual FC cards I need to change?
 
Joined
May 28, 2017
Messages
1
I still have the issue:

My freenas: Previous, when everything was working
FREENAS:9.10 Stable with QLE2464 (Fiberchannel adapter)
Vmware Esxi: 6.2 with QLE2462 (Fiberchannel adapter)

Upgrade:
FREENAS:11 Stable with QLE2464 (Fiberchannel adapter)
Vmware Esxi: 6.5 with QLE2462 (Fiberchannel adapter)

After upgrade, I see the following error

FREENAS:9.10 Stable with QLE2464 (Fiberchannel adapter)
Vmware Esxi: 6.2 with QLE2462 (Fiberchannel adapter)

May 28 10:34:07 nas ctlfeasync: WWPN 0x21000024ff386aec port 0x0000e8 path 3 target 0 arrived
May 28 10:34:07 nas ctlfeasync: WWPN 0x21000024ff386aed port 0x0000ef path 2 target 0 arrived
May 28 10:34:08 nas ctlfeasync: WWPN 0x21000024ff386aec port 0x0000e8 path 3 target 0 left
May 28 10:35:25 nas ctlfeasync: WWPN 0x2100001b329b5519 port 0x000001 path 0 target 0 arrived
May 28 10:35:25 nas ctlfeasync: WWPN 0x2101001b32bb5519 port 0x000001 path 1 target 0 arrived
May 28 10:36:15 nas isp3: Chan 255 LINK FAILED
May 28 10:36:16 nas ctlfeasync: WWPN 0x21000024ff386aec port 0x0000e8 path 3 target 0 arrived
May 28 10:36:16 nas isp2: Chan 255 LINK FAILED
May 28 10:37:46 nas isp0: Chan 255 LINK FAILED
May 28 10:37:48 nas isp1: Chan 255 LINK FAILED

I will be able to see the ISCSI HDD for some time then it goes offline.

To make it visible to Esxi I ended up switching the tunables as follows.
hint.isp.0.role from 0 to 1.

But still after some time the FC link goes down.
 
Joined
Aug 10, 2016
Messages
28
Hi, bumping up an old topic, but I wonder if you ever found a solution to this.
I'm new to FC and I am about to start learning, I've just bought the equipment.
And I'm doing my research before setting up everything ;)
 
Status
Not open for further replies.
Top