10:53 a.m.
hi all,
i ve 4 nodes running with current ovirt.
I ve only a problem on one host even after a fresh installation .
I ve installed the latest image.
Than I add the node to the cluster
Everything is working good.
After this I configure the network.
BUT, after a restart the host does not come up again.
I got this error: blk_cloned_rq_check_limits: over max size limit
every 5 seconds
I can continue with control-D
or I can login with root password to fix the problem. but i dont know what is the problem
and where does it came from
I ve also changed the sas disk to nvme storage, but I ve changed this on every host. And
this problem exists only on one host
i found this https://lists.centos.org/pipermail/centos/2017-December/167727.html
the output is
[root@kvm380 ~]# ./test.sh
Sys Block Node : Device max_sectors_kb
max_hw_sectors_kb
/sys/block/dm-0 : onn_kvm380-pool00_tmeta 256 4096
/sys/block/dm-1 : onn_kvm380-pool00_tdata 256 4096
/sys/block/dm-10 : onn_kvm380-var 256 4096
/sys/block/dm-11 : onn_kvm380-tmp 256 4096
/sys/block/dm-12 : onn_kvm380-home 256 4096
/sys/block/dm-13 : onn_kvm380-var_crash 256 4096
/sys/block/dm-2 : onn_kvm380-pool00-tpool 256 4096
/sys/block/dm-3 : onn_kvm380-ovirt--node--ng--4.3.7--0.20191121.0+1 256
4096
/sys/block/dm-4 : onn_kvm380-swap 256 4096
/sys/block/dm-5 : eui.0025385991b1e27a 512 2048
/sys/block/dm-6 : eui.0025385991b1e27a1 512 2048
/sys/block/dm-7 : onn_kvm380-pool00 256 4096
/sys/block/dm-8 : onn_kvm380-var_log_audit 256 4096
/sys/block/dm-9 : onn_kvm380-var_log 256 4096
cat: /sys/block/nvme0n1/device/vendor: Datei oder Verzeichnis nicht gefunden
/sys/block/nvme0n1: Samsung SSD 970 EVO 1TB 512 2048
/sys/block/sda : HP LOGICAL VOLUME 256 4096
is the nvme not starting correct
[root@kvm380 ~]# systemctl status multipathd
● multipathd.service - Device-Mapper Multipath Device Controller
Loaded: loaded (/usr/lib/systemd/system/multipathd.service; enabled; vendor preset:
enabled)
Active: active (running) since Di 2019-12-31 16:16:32 CET; 31min ago
Process: 1919 ExecStart=/sbin/multipathd (code=exited, status=0/SUCCESS)
Process: 1916 ExecStartPre=/sbin/multipath -A (code=exited, status=0/SUCCESS)
Process: 1911 ExecStartPre=/sbin/modprobe dm-multipath (code=exited, status=0/SUCCESS)
Main PID: 1921 (multipathd)
Tasks: 7
CGroup: /system.slice/multipathd.service
└─1921 /sbin/multipathd
Dez 31 16:47:58 kvm380.durchhalten.intern multipathd[1921]: nvme0n1: mark as failed
Dez 31 16:47:58 kvm380.durchhalten.intern multipathd[1921]: eui.0025385991b1e27a: Entering
recovery mode: max_retries=4
Dez 31 16:47:58 kvm380.durchhalten.intern multipathd[1921]: eui.0025385991b1e27a:
remaining active paths: 0
Dez 31 16:48:02 kvm380.durchhalten.intern multipathd[1921]: 259:0: reinstated
Dez 31 16:48:02 kvm380.durchhalten.intern multipathd[1921]: eui.0025385991b1e27a:
queue_if_no_path enabled
Dez 31 16:48:02 kvm380.durchhalten.intern multipathd[1921]: eui.0025385991b1e27a:
Recovered to normal mode
Dez 31 16:48:02 kvm380.durchhalten.intern multipathd[1921]: eui.0025385991b1e27a:
remaining active paths: 1
Dez 31 16:48:03 kvm380.durchhalten.intern multipathd[1921]: nvme0n1: mark as failed
Dez 31 16:48:03 kvm380.durchhalten.intern multipathd[1921]: eui.0025385991b1e27a: Entering
recovery mode: max_retries=4
Dez 31 16:48:03 kvm380.durchhalten.intern multipathd[1921]: eui.0025385991b1e27a:
remaining active paths: 0
why is it marked as failed?
if i create a new volume with cockpit and use it for bricks for gluster, every thing is
fine. until reboot
maybe some one can point me the direction