Stuart Rickards Posted April 10, 2017 Share Posted April 10, 2017 Hopefully someone can help with this issue. I am trying to configure a shared iscsi storage repository in xenserver 7.1, however when i try via the cli i get the target device is a fujitsu eternus DX60 s3 which is on the HCL SR_BACKEND_FAILURE_1 Error reporting error, unknown key DMP failed to activate mapper path, i get the same error via xen center. multipath -ll gives mpatha (3600000e00d290000002908f600000000) dm-1 FUJITSU ,ETERNUS_DXL size=18T features='1 queue_if_no_path' hwhandler='0' wp=rw `-+- policy='round-robin 0' prio=5 status=active |- 75:0:0:0 sdb 8:16 active ready running |- 79:0:0:0 sdc 8:32 active ready running |- 81:0:0:0 sde 8:64 active ready running `- 80:0:0:0 sdd 8:48 active ready running my multipath.conf file is as below # This configuration file is used to overwrite the built-in configuration of # multipathd. # For information on the syntax refer to `man multipath.conf` and the examples # in `/usr/share/doc/device-mapper-multipath-*/`. # To check the currently running multipath configuration see the output of # `multipathd -k"show conf"`. defaults { user_friendly_names yes queue_without_daemon no #find_multipaths yes } blacklist { devnode "^nvme.*" } # Leave this section in place even if empty blacklist_exceptions { } devices { device { vendor "FUJITSU" product "ETERNUS_DXL" prio alua path_grouping_policy group_by_prio path_selector "round-robin 0" failback immediate no_path_retry 10 path_checker tur dev_loss_tmo 2097151 fast_io_fail_tmo 1 } } Link to comment
1 Tobias Kreidl Posted April 10, 2017 Share Posted April 10, 2017 Was this entry already in the multipath.conf file or did you import it from some other source for your DX 60? -=Tobias 1 Link to comment
1 Alan Lantz Posted April 11, 2017 Share Posted April 11, 2017 Thinking along the same lines, you seem to be multipathing so maybe its just multipath.conf syntax ? --Alan-- 2 Link to comment
0 horizon limit Posted April 11, 2017 Share Posted April 11, 2017 Hi Stuart, This problem seems to be specific to the multipath i guess. If you disable the multipath, the SR shall be connected. Also, make sure to get the multipath settings from the vendor. HTH! Link to comment
0 Tobias Kreidl Posted April 11, 2017 Share Posted April 11, 2017 Did you perform an individual iscsiadm discovery of each path from each host? 1 Link to comment
0 Stuart Rickards Posted April 11, 2017 Author Share Posted April 11, 2017 Thanks for the responses, I have solved this issue by disabling the DMP driver and making a couple of adjustments to the multipath.conf file, the original multipath.conf file was apparently for the eternus DX60 s2 not the s3 version. Link to comment
0 Tobias Kreidl Posted April 11, 2017 Share Posted April 11, 2017 If you get some degree of connectivity, the multipath configuration file is generally the issue. Glad it's working for you now! -=Tobias Link to comment
0 Nishant Rapate Posted June 4, 2017 Share Posted June 4, 2017 HI Tobias, We are facing an strange issue in our xenserver 6.5 after expanding LUN size on storage, History - there are two xenservers in our pool and lenovo 1200s storage LUN is mapped with it, some days ago we were running out of space hence we had expanded space in the LUN but expanded space was not detecting by servers, so in order to get the expanded space detected on the server we tried to detach and attached SR (iscsi over LVM) again but no avail, expanded space was not detected so we tried to reboot the server in the sequence like first master and then slave but after rebooting VM's list is populated in the xencenter but when we try to start any vm which was created on the storage throws the error no boot disk found, also if we navigate the vms storage info it doesn't show vm partition path. but storage is connected properly to the server as we tried to create test vm in the same storage and we were able to create it. we ran these commands to unplug and plug in storage in the xenserver 1) Shut down all VMs on the SR. 2) # iscsiadm -m node -R 3) xe pbd-unplug uuid=<PBD UUID> 4) xe pbd-plug uuid=<PBD UUID> 5) Restart Hosts kindly suggest me how to get the vms back from the same storage without data loss as the vms are critical for business. Thanks Nishant Link to comment
0 Alan Lantz Posted June 5, 2017 Share Posted June 5, 2017 Nishant, Start a new thread. I don't think your post is related to MultiPath DMP. --Alan-- Link to comment
0 Ulrich Thomas Posted January 27, 2022 Share Posted January 27, 2022 Hello, I had the same error with a Nimble Storage device in XenServer 7.1CU2 I change "user_friendly_names" from "yes" to "no" in the multipath.conf file. That did the trick for me. Link to comment
Question
Stuart Rickards
Hopefully someone can help with this issue.
I am trying to configure a shared iscsi storage repository in xenserver 7.1, however when i try via the cli i get
the target device is a fujitsu eternus DX60 s3 which is on the HCL
SR_BACKEND_FAILURE_1
Error reporting error, unknown key DMP failed to activate mapper path,
i get the same error via xen center.
multipath -ll gives
mpatha (3600000e00d290000002908f600000000) dm-1 FUJITSU ,ETERNUS_DXL
size=18T features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=5 status=active
|- 75:0:0:0 sdb 8:16 active ready running
|- 79:0:0:0 sdc 8:32 active ready running
|- 81:0:0:0 sde 8:64 active ready running
`- 80:0:0:0 sdd 8:48 active ready running
my multipath.conf file is as below
# This configuration file is used to overwrite the built-in configuration of
# multipathd.
# For information on the syntax refer to `man multipath.conf` and the examples
# in `/usr/share/doc/device-mapper-multipath-*/`.
# To check the currently running multipath configuration see the output of
# `multipathd -k"show conf"`.
defaults {
user_friendly_names yes
queue_without_daemon no
#find_multipaths yes
}
blacklist {
devnode "^nvme.*"
}
# Leave this section in place even if empty
blacklist_exceptions {
}
devices {
device {
vendor "FUJITSU"
product "ETERNUS_DXL"
prio alua
path_grouping_policy group_by_prio
path_selector "round-robin 0"
failback immediate
no_path_retry 10
path_checker tur
dev_loss_tmo 2097151
fast_io_fail_tmo 1
}
}
Link to comment
9 answers to this question
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now