Article ID: 121514, created on May 9, 2014, last review on Apr 13, 2015

  • Applies to:
  • Virtuozzo 6.0

Symptoms

prlctl commands hang and then fail with error when timeout is reached.

# prlctl list -a
Login failed: Unable to connect to the server "localhost". An invalid response has been received from the server. Make sure that Parallels Server is running on the server "localhost" and is up to date and try again.

Cause

This issue might happen if some filesystem, where VMs or VM templates are registered, has become inaccessible, operations (such as df) with its mount point hang endlessly:

[root@PCS1 ~]# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/sda3             394G   10G  364G   3% /
tmpfs                  32G  5.2M   32G   1% /dev/shm
/dev/sda1             2.0G  122M  1.8G   7% /boot
/dev/mapper/vg_pcs1-lv_pstorage
                      3.2T  467G  2.6T  16% /pstorage
/dev/mapper/vg_pcs1-lv_vz
                    40G  9.4G   29G  25% /vz
/pstorage/cluster-mds/DiskDescriptor.xml
                    9.9G  335M  9.1G   4% /pstorage/cluster-mds/ploop_mds.img.mnt
^C

Usually it's NFS file system which stores VM templates. NFS mount point tend to be the most vulnerable and fragile:

[root@PCS1 ~]# mount | grep nfs
1.2.3.4:/mnt/cluster/PCS1 on /templates type nfs (rw,timeo=14,intr,addr=1.2.3.4)

In such cases you will find nfs-related errors in dmesg:

[root@PCS1 ~]# dmesg | grep nfs
[10405249.292138] ct0 nfs: server 1.2.3.4 not responding, still trying
[10405259.798122] ct0 nfs: server 1.2.3.4 not responding, still trying

Resolution

Find the reason why operations with this mount point hang, fix the root cause. Often it's due to inaccessible NFS server.

Search Words

Unable to connect to the server "localhost"

prlctl hang

c62e8726973f80975db0531f1ed5c6a2 2897d76d56d2010f4e3a28f864d69223 0dd5b9380c7d4884d77587f3eb0fa8ef

Email subscription for changes to this article
Save as PDF