cpjitservices Posted March 22, 2017 Posted March 22, 2017 Hi all We have upgraded a few of Our Dell Servers R510/710. NICs are having issues where by the network is slow and drops out. All drivers up to date according to Dell Update tool and Dells website. NICS are QLogic BCM5716C. Its not the network links themselves as a 2012 Server will sit on the same link all day long without issues, seems to be a 2016 problem. The weird thing is, I can ping the VM's on the Host and they ping continuously without hitch, ping the host and ping will drop out, reply, reply, dropout etc. Anyone come across it ? Thanks in advance all.
ITGURU Posted March 22, 2017 Posted March 22, 2017 Are you using the built in the teaming with server 2016? Should use it's own drivers, or get them from the QLOGIC site? I've moved one of our non critical servers to 2016 and was able to use the 2012 R2 drivers rather than the defaults and they have been fine.
cpjitservices Posted March 22, 2017 Author Posted March 22, 2017 Are you using the built in the teaming with server 2016? Should use it's own drivers, or get them from the QLOGIC site? I've moved one of our non critical servers to 2016 and was able to use the 2012 R2 drivers rather than the defaults and they have been fine. Thanks, as far as I'm aware we are NOT using teaming... How would I tell ? This is literally just a boggo build of 16 with HyperV with Virtual Switch. Pinging a VM on the host now and doesnt miss a beat. Ping the Host and its request timed out, then replies, the request timed out and so on. I'm also RDP'd into the Host and Its disconnecting me constantly.
ITJS2015 Posted March 22, 2017 Posted March 22, 2017 Where are using RDP ? from the physical server to your VM or from a laptop to a vm ? Test your connections on hardwire and also wireless with any devices which you can connect to your vm
Blue_Cookeh Posted March 22, 2017 Posted March 22, 2017 (edited) Have you tried disabling VMQ in the NIC properties in device manager? I stick to Intel NICs since they seem to be the most stable and supported across all OSes, using them with Hyper-V on Dell servers without this issue but when we did have other NICs the link above solved a lot of issues. Edited March 22, 2017 by Blue_Cookeh
cpjitservices Posted March 22, 2017 Author Posted March 22, 2017 (edited) Where are using RDP ? from the physical server to your VM or from a laptop to a vm ? Test your connections on hardwire and also wireless with any devices which you can connect to your vm RDP to physical host from different Subnet. Put a Server 2012 R2 on the same link and no issues. Also no issues on the VM's on that 2016 host. Edited March 22, 2017 by cpjitservices
cpjitservices Posted March 22, 2017 Author Posted March 22, 2017 (edited) Have you tried disabling VMQ in the NIC properties in device manager? I stick to Intel NICs since they seem to be the most stable and supported across all OSes, using them with Hyper-V on Dell servers without this issue but when we did have other NICs the link above solved a lot of issues. Not looked at that - I dont see it on the NIC properties. **Found it** Will test. Edited March 22, 2017 by cpjitservices
cpjitservices Posted March 22, 2017 Author Posted March 22, 2017 Nope disabling VMQ has made no difference at all. Tried 2012 driver. Same Issues Guess it's a Windows 16 bug.
fiza Posted March 22, 2017 Posted March 22, 2017 Have you tried disabling VMQ in the NIC properties in device manager? I stick to Intel NICs since they seem to be the most stable and supported across all OSes, using them with Hyper-V on Dell servers without this issue but when we did have other NICs the link above solved a lot of issues. We had this issue with Broadcom NICS. By the part number it sounds link your are Broadcom based too (BCM). https://notesbytom.wordpress.com/2015/07/27/broadcom-qlogic-vmq-hyper-v-network-bug/
ITJS2015 Posted March 22, 2017 Posted March 22, 2017 I'm not to sure if it has anything to do with your firewall settings, but I had a RDP session drop out a few months back which the problems was a tick box on firwall settings under RDP with it being a Public or guest network.
cpjitservices Posted March 22, 2017 Author Posted March 22, 2017 I'm not to sure if it has anything to do with your firewall settings, but I had a RDP session drop out a few months back which the problems was a tick box on firwall settings under RDP with it being a Public or guest network. Checked that PFsense not blocking it.. Everything drops out - Screen Connect is doing the same as well as ping.
cpjitservices Posted March 22, 2017 Author Posted March 22, 2017 We have a result. A bizarre one too. But everything is as it should be....
cpjitservices Posted March 22, 2017 Author Posted March 22, 2017 Another server has, as it turns out a dodgy NIC port. Other server also running Server 2016. Maybe thats the link perhaps ? as no other server is affected but these are the only 2 on 16. Anyway, through troubleshooting I Shutdown my other server and pings returned to normal on the server that was misbehaving and immediately thought thats odd. RDP and Screen Connect also worked perfect. Sure enough, turned other server on and as soon as it started posting, pings, rdp etc dropped out on the naughty server. Unplugged cable from port 1 on back of server and plugged into port 2 and everything is now working as it should be. Odd.
Norphy Posted March 22, 2017 Posted March 22, 2017 Odd but something similar happened to me a couple of weeks back. The entire network across all of our sites was having issues, constant drop outs, slowdowns, the works. We traced it to the core switch in one of our buildings. We did everything up to and including replacing the core with something more modern. Still things weren't right. The issue turned out to be the wireless link to one of the primaries. Remove the wireless link, everything was happy. Plug it back in and we got dropouts, missed pings, the works. The wireless link got replaced and the federation started working again 1
cpjitservices Posted March 22, 2017 Author Posted March 22, 2017 (edited) Odd but something similar happened to me a couple of weeks back. The entire network across all of our sites was having issues, constant drop outs, slowdowns, the works. We traced it to the core switch in one of our buildings. We did everything up to and including replacing the core with something more modern. Still things weren't right. The issue turned out to be the wireless link to one of the primaries. Remove the wireless link, everything was happy. Plug it back in and we got dropouts, missed pings, the works. The wireless link got replaced and the federation started working again Bizarre could understand it if it was a switch port, next time im in the server room ill plug the link back into port 1 on the server and see when the pings drop if the lights go out on the NIC. Might confirm if its hardware. These things are sent to try us and resignation letter can go back in the drawer lol. Strange how it only affected this server, though the server it was affecting is the same make/model as the troublesome one and on the same OS... the rest of the farm is a different model and on 2012 R2. Perhaps they had just taken offence to each other lol... bullying in the server farm. Edited March 22, 2017 by cpjitservices
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now