Unable to locate target iqn iqn.2016-05.com.petasan:00003 in storeage node
phanvlong
9 Posts
March 23, 2022, 9:03 amQuote from phanvlong on March 23, 2022, 9:03 amHi all,
My cluster have 4 node: 2 manager nodes with 4 OSDs, 1 manager without OSD run Management and Monitoring Services only, 1 storage node with 3 OSDs, I created iSCSI disks and it run smoothly but there 2 problems:
- Ceph Heath warning with message:
- Node console alway show message: Unable to locate target iqn iqn.2016-05.com.petasan:00003 in storeage node
How can I fix this
Thank for your help
Phan Long
Hi all,
My cluster have 4 node: 2 manager nodes with 4 OSDs, 1 manager without OSD run Management and Monitoring Services only, 1 storage node with 3 OSDs, I created iSCSI disks and it run smoothly but there 2 problems:
- Ceph Heath warning with message:
- Node console alway show message: Unable to locate target iqn iqn.2016-05.com.petasan:00003 in storeage node
How can I fix this
Thank for your help
Phan Long
admin
2,930 Posts
March 23, 2022, 10:27 amQuote from admin on March 23, 2022, 10:27 amThe PG per OSD issue : the current current value is too low. You probably had created several default pools during deployment then deleted some, so end up with pool(s) with low PGs. if you use version 3 (or upgraded to it) you can enable pg autoscale on the pool, or if this is a test cluster, create a new pool with autoscale on to avoid data rebalance.
The iSCSI message is likely you had a previous disk that was deleted (or from a prev cluster install) that is still configured in a client machine and is trying to connect to no longer existence disk, the server logs that the iqn does not exists, you need to update your client iSCSI config.
The PG per OSD issue : the current current value is too low. You probably had created several default pools during deployment then deleted some, so end up with pool(s) with low PGs. if you use version 3 (or upgraded to it) you can enable pg autoscale on the pool, or if this is a test cluster, create a new pool with autoscale on to avoid data rebalance.
The iSCSI message is likely you had a previous disk that was deleted (or from a prev cluster install) that is still configured in a client machine and is trying to connect to no longer existence disk, the server logs that the iqn does not exists, you need to update your client iSCSI config.
phanvlong
9 Posts
March 24, 2022, 9:18 amQuote from phanvlong on March 24, 2022, 9:18 am@admin: Thank for your reply, I fixed the PG per OSD issue, can you tell me how to find out which client is trying to connect to does not exit IQN
Best regards
Phan Long
@admin: Thank for your reply, I fixed the PG per OSD issue, can you tell me how to find out which client is trying to connect to does not exit IQN
Best regards
Phan Long
admin
2,930 Posts
March 24, 2022, 10:50 amQuote from admin on March 24, 2022, 10:50 amyou can check iscsi configuration on your client machines. or you can increase iSCSI logging on PetaSAN
# turn on dynamic debug log
echo "file drivers/target/iscsi/iscsi_target_login.c +pf" > /sys/kernel/debug/dynamic_debug/control
dmesg | grep -A 5 "Received iSCSI login request from"
# turn off log
echo "file drivers/target/iscsi/iscsi_target_login.c -pf" > /sys/kernel/debug/dynamic_debug/control
you can check iscsi configuration on your client machines. or you can increase iSCSI logging on PetaSAN
# turn on dynamic debug log
echo "file drivers/target/iscsi/iscsi_target_login.c +pf" > /sys/kernel/debug/dynamic_debug/control
dmesg | grep -A 5 "Received iSCSI login request from"
# turn off log
echo "file drivers/target/iscsi/iscsi_target_login.c -pf" > /sys/kernel/debug/dynamic_debug/control
phanvlong
9 Posts
March 25, 2022, 3:30 amQuote from phanvlong on March 25, 2022, 3:30 am@admin: I found out the client, Thank you very much
Best regards
Phan Long
@admin: I found out the client, Thank you very much
Best regards
Phan Long
Unable to locate target iqn iqn.2016-05.com.petasan:00003 in storeage node
phanvlong
9 Posts
Quote from phanvlong on March 23, 2022, 9:03 amHi all,
My cluster have 4 node: 2 manager nodes with 4 OSDs, 1 manager without OSD run Management and Monitoring Services only, 1 storage node with 3 OSDs, I created iSCSI disks and it run smoothly but there 2 problems:
- Ceph Heath warning with message:
- Node console alway show message: Unable to locate target iqn iqn.2016-05.com.petasan:00003 in storeage node
How can I fix this
Thank for your help
Phan Long
Hi all,
My cluster have 4 node: 2 manager nodes with 4 OSDs, 1 manager without OSD run Management and Monitoring Services only, 1 storage node with 3 OSDs, I created iSCSI disks and it run smoothly but there 2 problems:
- Ceph Heath warning with message:
- Node console alway show message: Unable to locate target iqn iqn.2016-05.com.petasan:00003 in storeage node
How can I fix this
Thank for your help
Phan Long
admin
2,930 Posts
Quote from admin on March 23, 2022, 10:27 amThe PG per OSD issue : the current current value is too low. You probably had created several default pools during deployment then deleted some, so end up with pool(s) with low PGs. if you use version 3 (or upgraded to it) you can enable pg autoscale on the pool, or if this is a test cluster, create a new pool with autoscale on to avoid data rebalance.
The iSCSI message is likely you had a previous disk that was deleted (or from a prev cluster install) that is still configured in a client machine and is trying to connect to no longer existence disk, the server logs that the iqn does not exists, you need to update your client iSCSI config.
The PG per OSD issue : the current current value is too low. You probably had created several default pools during deployment then deleted some, so end up with pool(s) with low PGs. if you use version 3 (or upgraded to it) you can enable pg autoscale on the pool, or if this is a test cluster, create a new pool with autoscale on to avoid data rebalance.
The iSCSI message is likely you had a previous disk that was deleted (or from a prev cluster install) that is still configured in a client machine and is trying to connect to no longer existence disk, the server logs that the iqn does not exists, you need to update your client iSCSI config.
phanvlong
9 Posts
Quote from phanvlong on March 24, 2022, 9:18 am@admin: Thank for your reply, I fixed the PG per OSD issue, can you tell me how to find out which client is trying to connect to does not exit IQN
Best regards
Phan Long
@admin: Thank for your reply, I fixed the PG per OSD issue, can you tell me how to find out which client is trying to connect to does not exit IQN
Best regards
Phan Long
admin
2,930 Posts
Quote from admin on March 24, 2022, 10:50 amyou can check iscsi configuration on your client machines. or you can increase iSCSI logging on PetaSAN
# turn on dynamic debug log
echo "file drivers/target/iscsi/iscsi_target_login.c +pf" > /sys/kernel/debug/dynamic_debug/controldmesg | grep -A 5 "Received iSCSI login request from"
# turn off log
echo "file drivers/target/iscsi/iscsi_target_login.c -pf" > /sys/kernel/debug/dynamic_debug/control
you can check iscsi configuration on your client machines. or you can increase iSCSI logging on PetaSAN
# turn on dynamic debug log
echo "file drivers/target/iscsi/iscsi_target_login.c +pf" > /sys/kernel/debug/dynamic_debug/control
dmesg | grep -A 5 "Received iSCSI login request from"
# turn off log
echo "file drivers/target/iscsi/iscsi_target_login.c -pf" > /sys/kernel/debug/dynamic_debug/control
phanvlong
9 Posts
Quote from phanvlong on March 25, 2022, 3:30 am@admin: I found out the client, Thank you very much
Best regards
Phan Long
@admin: I found out the client, Thank you very much
Best regards
Phan Long