Hello everyone,
I have a centos virtual server in Workstation 14 (installed on windows server 2016).
I have unexplained periodic holes during the day, which last up to half an hour and can only be bypassed by restarting the linux server (ora waiting a lot).
The services related to the linux virtual server are disconnected and it becomes impossible to ping. This creates continuous work interruptions for me.
In these moments the VM Workstation becames often not responsive. Sometimes I've only to shutdown the windows physical server to restart the centos VM
Within the network I am monitoring the various machines, virtual and otherwise, via pinginfoview: the windows server can be pinged without any problems, as well as other physical machines in the network.
I thought it was a problem of the aforementioned linux server, then I turned on another virtual linux server (proxmox) on the Vm workstation and I realized that the ping problems are almost always simultaneous in the two active virtual servers.
All this after having:
completely uninstalled webroot
disabled windows defender and windows firewall.
NB: from the vmware log I have not noticed particular coincidences
Now I don't really know what other tests to do ...
Tips?
Thank you,
Gianluca
Hi,
Which version of Centros did you install?
ARomeo
Ciao Romeo
this is the version: CentOS Linux release 7.7.1908
Tks,
gianluca
Ok!
Centos 7.x is in the compatibility matrix for vmware workstation 14
Have you installed the vmware tools?
Yes, installed
Hi,
Are you perhaps using autoprotect snapshots?
--
Wil
Hi Wila,
no, the autoprotect snapshot is disabled.
G.
HI,
OK.
Can you please attach the vmware.log file(s) of that VM to a reply here and also tell me how much RAM your host has?
Note that the button to attach files is at the bottom right of the reply window.
--
Wil
Hi,
Your host CPU has 4 physical cores and 8 threads (see Intel® Xeon® Processor E3-1230 v6 (8M Cache, 3.50 GHz) Product Specifications )
You assigned the guest with 4 vCPU cores. So my best guess is that you have over-provisioned the guest CPU core count (threads don't count).
Drop the guest CPU core count to 2 cores.
The assigned amount of RAM _should_ be OK, but for testing I also suggest to halve it as I do not know how much you tax the host with other things.
IOW change guest RAM to 8 GB.
Try and see if the problem still occurs.
--
Wil
NB: At the moment, having the need not to create further inconvenience to the users of the program (data loss and interruptions), I am trying to transfer the wmdk file to vhdx, using a W10pro pc as Hyper-v server.
At the moment I have not yet figured out if it will work.
Given my great difficulties with VM Workstation, I wanted to switch, as soon as the quarantine ends, to Vm exsi.
I ask you: the Vmdk to Vhdx conversions, then move on to Vhdx to Vm esxi could create problems and compromise the virtual files preventing me from concluding the various steps with success?
Many tks and regards,
G.
Ok, I wil try so and let you know.
Thank you,
Gianluca
nothing to do, I configured the VM as you told me, I turned the centos back on and after a while I had another 20 minute hole (from 16.08 to 16.27). I'll re-attach the VmWare log. Thanks
@wila
Hi,
Ok, so that wasn't likely to be triggering your problem then.
Not seeing anything happen in that time frame.
The last thing that happens -quite a while before your issue- is that a Bluetooth connection can not be made.
You might want to disable bluetooth for the VM. As it's a server, chances are likely that you aren't using that feature anyways.
If that doesn't help... then try removing the USB controller as well.
--
Wil
Hi,
Might work, don't know really as I'm not a Hyper-V user...
--
Wil
After several tests on the Vmware workstation, the problem is still present and I have absolutely no idea where it comes from.
Basically:
1) I made all the changes suggested in this thread
2) I have cloned the centos machine on virtualbox and it seems to work
3) I created a centos vmdk test machine on vmware and it also has periodic holes, although often not coinciding with the other
4) I checked all the windows logs but I did not find any coincidence between the holes and any errors or the start of windows automations.
The only singular datum is that the ping holes almost always start at precise hours (These are the last holes; 08/04/2020 12:00:17 Request Timeout
; 08/04/2020 06:00:48 Request Timeout; 08/04/2020 02:10:13 Request Timeout; 07/04/2020 23:00:23 Request Timeout) even if the times are difficult to repeat the following day.
Could this be an internal problem in Centos? Maybe some applications that start automatically at certain times and cause the VMware tilt?
Thanks,
G.
Hi,
If it happens at regular times then that certainly is suspicious.
I would check what happens at the host on those times.
For example I have seen reports in the past about responsiveness issues appearing at regular times which -after some time- turned out to coincide with the backup schedule of the host.
Could it be that the host runs something around those same times?
For example a disk defrag or something similar?
--
Wil