Multipath-tools Errors in PVE 6.3 Updated

[Rais Ahmed]

Hi,

I have install and configure 3 Nodes Cluster Proxmox Virtual Environment 6.3-3 with multipath-tools Installed: 0.7.9-3+deb10u1 and seeing following errors continuously.

multipathd[10763]: get_alua_info: get_asymmetric_access_state returned -3
multipathd[10763]: sdi: couldn't get asymmetric access state
multipathd[10763]: get_alua_info: get_asymmetric_access_state returned -3
multipathd[10763]: sdq: couldn't get asymmetric access state
multipathd[10763]: get_alua_info: get_asymmetric_access_state returned -3
multipathd[10763]: sdg: couldn't get asymmetric access state
multipathd[10763]: get_alua_info: get_asymmetric_access_state returned -3
multipathd[10763]: sdp: couldn't get asymmetric access state
multipathd[10763]: get_alua_info: get_asymmetric_access_state returned -3
multipathd[10763]: sdf: couldn't get asymmetric access state
multipathd[10763]: get_alua_info: get_asymmetric_access_state returned -3
multipathd[10763]: sdo: couldn't get asymmetric access state
multipathd[10763]: get_alua_info: get_asymmetric_access_state returned -3
multipathd[10763]: sdh: couldn't get asymmetric access state
multipathd[10763]: get_alua_info: get_asymmetric_access_state returned -3
multipathd[10763]: sdn: couldn't get asymmetric access state

Why I am seeing these errors?
Does it have any impact to performance/stability?
How to resolve it?

Please help.

 

[Rais Ahmed]

Anyone please help...!

 

[Rais Ahmed]

Proxmox team...?

 

[2707974]

Hi,

do you add some LUN from storage?
Can you please give us output from commands

multipath -ll

dmsetup ls

dmsetup deps

 

[Rais Ahmed]

Hi Thank you for response.

Yes, LUNs connected via iSCSI configured in multipath.

multipath -ll
Jan 21 22:44:58 | get_alua_info: get_asymmetric_access_state returned -3
Jan 21 22:44:58 | sdb: couldn't get asymmetric access state
Jan 21 22:44:58 | get_alua_info: get_asymmetric_access_state returned -3
Jan 21 22:44:58 | sdc: couldn't get asymmetric access state
Jan 21 22:44:58 | get_alua_info: get_asymmetric_access_state returned -3
Jan 21 22:44:58 | sdd: couldn't get asymmetric access state
Jan 21 22:44:58 | get_alua_info: get_asymmetric_access_state returned -3
Jan 21 22:44:58 | sde: couldn't get asymmetric access state
mpath0 (36e0008410036ab762422364a00000008) dm-5 HUAWEI,XSG1
size=5.0T features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=-1 status=active
  |- 1:0:0:1 sdb 8:16  active ready running
  |- 2:0:0:1 sdc 8:32  active ready running
  |- 3:0:0:1 sdd 8:48  active ready running
  `- 4:0:0:1 sde 8:64  active ready running
Jan 21 22:44:58 | get_alua_info: get_asymmetric_access_state returned -3
Jan 21 22:44:58 | sdo: couldn't get asymmetric access state
Jan 21 22:44:58 | get_alua_info: get_asymmetric_access_state returned -3
Jan 21 22:44:58 | sdp: couldn't get asymmetric access state
Jan 21 22:44:58 | get_alua_info: get_asymmetric_access_state returned -3
Jan 21 22:44:58 | sdn: couldn't get asymmetric access state
Jan 21 22:44:58 | get_alua_info: get_asymmetric_access_state returned -3
Jan 21 22:44:58 | sdq: couldn't get asymmetric access state
mpatha (36e0008410036ab7624bd8f6c00000009) dm-13 HUAWEI,XSG1
size=500G features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=-1 status=active
  |- 1:0:0:4 sdo 8:224 active ready running
  |- 2:0:0:4 sdp 8:240 active ready running
  |- 3:0:0:4 sdn 8:208 active ready running
  `- 4:0:0:4 sdq 65:0  active ready running

dmsetup ls
Huawei5TB-vm--201--disk--0      (253:7)
mpath0-part1    (253:6)
mpath0  (253:5)
Huawei--1TB-vm--804--disk--0    (253:12)
pve-data_tdata  (253:3)
pve-data_tmeta  (253:2)
mpatha  (253:13)
pve-swap        (253:0)
pve-root        (253:1)
Huawei--1TB-vm--101--disk--2    (253:11)
pve-data        (253:4)
Huawei--300GB-vm--702--disk--0  (253:8)
Huawei--1TB-vm--101--disk--1    (253:10)
mpatha-part1    (253:14)
Huawei--1TB-vm--101--disk--0    (253:9)

dmsetup deps
Huawei5TB-vm--201--disk--0: 1 dependencies      : (253, 6)
mpath0-part1: 1 dependencies    : (253, 5)
mpath0: 4 dependencies  : (8, 64) (8, 48) (8, 32) (8, 16)
Huawei--1TB-vm--804--disk--0: 1 dependencies    : (8, 145)
pve-data_tdata: 1 dependencies  : (8, 3)
pve-data_tmeta: 1 dependencies  : (8, 3)
mpatha: 4 dependencies  : (65, 0) (8, 208) (8, 240) (8, 224)
pve-swap: 1 dependencies        : (8, 3)
pve-root: 1 dependencies        : (8, 3)
Huawei--1TB-vm--101--disk--2: 1 dependencies    : (8, 145)
pve-data: 2 dependencies        : (253, 3) (253, 2)
Huawei--300GB-vm--702--disk--0: 1 dependencies  : (8, 81)
Huawei--1TB-vm--101--disk--1: 1 dependencies    : (8, 145)
mpatha-part1: 1 dependencies    : (253, 13)
Huawei--1TB-vm--101--disk--0: 1 dependencies    : (8, 145)

 

[2707974]

Do you restart servers, if you can, after adding LUNs?
You have EMULEX HBA?
You use both LUNs

mpath0-part1

and

mpatha-part1

Problem is in partition table.
You can reload multipath

systemctl reload multipathd

This will reload config from

/etc/multipath.com
/etc/multipath/wwid

 

[Rais Ahmed]

Hi,
Already done mentioned items,
yes using both LUNs
configuration reloaded.
Please help.

 

[2707974]

Hi,
Already done mentioned items,
yes using both LUNs
configuration reloaded.
Please help.

Hi,

My advice is to restart one node, you say have it three.
If that solve problem you can migrate VM and restart second ...

If that do not resolve problem, only you can do is to "delete" problematic disk's from multipath

multipathd del path sdb
echo 1 > /sys/block/sdb/device/delete

To this two command for each problematic disk in environment [sdb, sdc, sdd, sde].