Hi everybody,
I have an x86 machine (on VMWare 4) that sometime crashes. I've already tested different router os version (5.19 too) but with no success.
I have another x86 machine working fine with 4.17.
Any idea? The only weird thing is this IRQ sharing resources (see image) for 2 ether interfaces... Is it normal?

Thanks!

I have the exact same thing with an x86 on ESXi 4.1 with an i7 2600k it was randomly crashing and showing 100% cpu usage in vmware.

I can assure that is NOT a problem from the IRQ sharing...
I removed the interface, but server still reboot...
Any answer from MikroTik guys? Any suggestion?

Thanks again...

What cards do you have in the machine? I had a i2600 doing the same thing on 5.1, however I stopped the router and change the network adapters in ESXi 4.1.0 from flexible to e1000 (cards I'm using) and the resources went from up near 80-90% down to 20-30%.

What cards do you have in the machine? I had a i2600 doing the same thing on 5.1, however I stopped the router and change the network adapters in ESXi 4.1.0 from flexible to e1000 (cards I'm using) and the resources went from up near 80-90% down to 20-30%.

I had 3 VM with ROS. 2 of them with ROS 4.17, 4 ethernet connection and no problems at all...
1 VM was working with ROS 5.x (I tried 5.12, 5.14, 5.18) but sometimes it crashes.
Now i downgraded to 4.17... Lets see how it works...

On every VM, NIC are e1000...

Bye!

No one from MikroTik?
Thanks...

Same problem here, with Vmware 5.1 and Intel core i7

Same problem here, with Vmware 5.1 and Intel core i7

RouterOS version?
With 4.17 no problem at all...

We have the same problem (vmware crashing) using vmware 5.1 and routeros 5.20 / 6.0rc (tested):
- NOT_IMPLEMENTED bora/vmkernel/sched/memsched.c 17724
- Exception 14

vmware system crash can be reproduced with mikrotik:
- /system/shutdown => crash
- import certificates => crash
- do some random stuff => crash

Seems to be a problem of vmware 5.1 since the same routeros doesn't make any problems on vmware 5.0.1. Too bad on the new machine we can't downgrade anymore...

Any known workaround on vmware 5.1 available ? Did someone find a stable configuration?

I have now installed ROS 4.17 and no crashes in vmware.
Maybe someone from mikrotik could take a look on this issue in 5.x/6.x...

I can assure that is NOT a problem from the IRQ sharing...
I removed the interface, but server still reboot...
Any answer from MikroTik guys? Any suggestion?

Thanks again...

Had same problem with xeon multi CPU x86
in 5.18 MT solve the problem try it (but if u using MT on vmware dont know)


disable firewall see if it helps
also if this dont help just disable multicpu

I can assure that is NOT a problem from the IRQ sharing...
I removed the interface, but server still reboot...
Any answer from MikroTik guys? Any suggestion?

Thanks again...

Had same problem with xeon multi CPU x86
in 5.18 MT solve the problem try it (but if u using MT on vmware dont know)


disable firewall see if it helps
also if this dont help just disable multicpu

This is not a great "workaround" don't you think...? I'm sorry but I need both firewall and multicore...
I hope in newest release of routeros this problems will be fixed...

Thanks anyway...

hope so
don't forget write to support and send supout.rif

Today vmware crash again caused by mikrotik routeros (now 4.17). Seems this also doesn't work stable...
Really too bad mikrotik seems to doesn't care about vmware environments since most current version also causing crashes in vmware 5.1.
Maybe it's a pure vmware issue, if so i hope it will be fixed in next patch.

Today vmware crash again caused by mikrotik routeros (now 4.17). Seems this also doesn't work stable...
Really too bad mikrotik seems to doesn't care about vmware environments since most current version also causing crashes in vmware 5.1.
Maybe it's a pure vmware issue, if so i hope it will be fixed in next patch.

In this case the only thing I can say is that we have 5 RouterOS Virtual Machines (on ESXi 4.1) with ROS 4.17 and no problem at all.
Uptime are up 60 or 70 days without problems (with more than 100 Mbit per VM of traffic per second).

In your case, could be another problem...

Regards...

I wrote about vmware 5.1, not 4.1.
We have another vmware machine running with 5.0 and there are no problems at all until now.
So it's something with mikrotik routeros and vmware esxi 5.1.

I wrote about vmware 5.1, not 4.1.
We have another vmware machine running with 5.0 and there are no problems at all until now.
So it's something with mikrotik routeros and vmware esxi 5.1.

We have 6 virtual machines running ROS in esxi 5.0 u1 with ROS 5.18, uptime 61days on all of them as well.

Have not tried on 5.1 yet, but if its the host crashing, that sounds like a problem on VMWare's side.

Sure, i didn't say its mikrotiks fault. But also it only happens with mikrotik machines, so there could be some special memory or cpu access in routeros causing vmware to crash.
So, at least this thread can be used to warn all mikrotik users before installing esx 5.1.

I know this is somewhat off topic but would you guys mind sharing some information about your setups ? I'm looking to move my current system to a VMWare environment. Its a fairly small setup with under 100 users, the server itself is 1.7GHZ, 256mb ram and resource usage barely crosses 10%. I'm doing load balancing on 3x adsl lines with each "wan" having its own 100mb ethernet card. I'd like to be able to setup a web host, proxy and other services through vmware to consolidate it all on one box for a low cost solution.

What kind of OS are you guys running, what other services are provided through VMWare (like proxy?), whats the system configuration and most importantly whats the reboot time like ? The area I'm in has frequent power outs and with a hardware based MK server I can get a reboot time of about 30 seconds.

You can reply in my topic to avoid dragging this thread further offtopic

Hi everybody,
I have an x86 machine (on VMWare 4) that sometime crashes. I've already tested different router os version (5.19 too) but with no success.
I have another x86 machine working fine with 4.17.
Any idea? The only weird thing is this IRQ sharing resources (see image) for 2 ether interfaces... Is it normal?

Thanks!

i have many ESXI 5.0 server
at dell R210II(XEON E3 1220) ROUTEROS 5.X crash(Every three days a crash) .but routeros 4.11 run perfect


at dell r710 (xeon 55xx) ROUTEROS 5.X run perfect

at i3 550 & i5 750 routeros 5.18 run perfect

4.11 at esxi5.0

5.7 esxi 4.1