When I added an extra SCSI card to my ESX 3.5 test system I could not get it to reconnect again with VirtualCenter. After a close inspection I noticed that vmnic0 did not exist anymore(esxcfg-nics -l), it got renamed to vmnic1. Because of that my vSwitch0 did not have a NIC linked to it anymore. I linked the new vmnic1 to my vSwitch0(esxcfg-vswitch vSwitch0 -L vmnic1) and everything was up and running again. I’m glad this didn’t happen in a production environment with several vSwitches and portgroups, than again it is possible to rename them again by manually editing the /etc/vmware/esx.conf file. There’s a knowledge base article on this behavior on the VMware support site, I noticed it a couple of weeks ago an now I witnessed it. Definitely not the first place I would normally look. I hope VMware can fix this glitch, although I can fully understand why this happens.
Sponsors
Recent Comments
- William Lam on KB article about SvMotion / VDS / HA problem republished with script to mitigate!
- William Lam on KB article about SvMotion / VDS / HA problem republished with script to mitigate!
- Erick on Multiple-NIC vMotion in vSphere 5…
- Diego Cogo on KB article about SvMotion / VDS / HA problem republished with script to mitigate!
- Bilal Hashmi on KB article about SvMotion / VDS / HA problem republished with script to mitigate!
Tags
2.5
3.0.x
3.5
4.1
5
5.0
BC-DR
Blogging
Book
Bugs
cloud
design
drs
ESX
esxi
ha
Howto
linux
networking
News
patches
performance
powershell
Scripting
scripts
security
service console
srm
Storage
Tools
training
update
upgrade
VCB
vcdx
vcenter
vcloud
vdi
VirtualCenter
vmtn
VMware
vmworld
vSphere
vstorage
whitepaper
Disclaimer
The views expressed anywhere on this site are strictly mine and not the opinions and views of VMware.
Idd bekend verhaal, dit gebeurt al sinds 1875 A.D., nou ja…ik ken het al vanaf versie 2.5.x. Ik vraag me af of ze hier ooit iets aan gaan doen. Wordt nog leuker als je meerdere netwerkkaarten in je systeem hebt, dan nummert nl. alles zo ongeveer door :S.
Trouwens…VMs hebben precies dit zelfde gedrag:).
Serieus? Ik heb er gelukkig nog nooit last van gehad. En zeker als je 6 nics hebt zul je hier niet vrolijk van worden. Van VM’s is het een bekend verschijnsel ja… Zeker in de beginfase ging het vaak fout als er een nieuwe tools uit was.
Interesting, I actually just ran into this while adding an HBA to a test server not an hour ago. Was going to call it a night and pick up looking into the issue in the morning. Glad I like your site! Will save me some hassle!!
Heb het idd een keer gehad met 7 NICs of zoiets…was erg grappig maar niet heus! Dan is het echt van…welke nic hangt aan welk fysieke netwerk. Maar goed, zo blijft er werk….en dat is goed!!
I have an ESX host 2.5.4 that is disconnected and greyed out in Virtual Center. In order to try to reconnect it, first I must execute the “disconnect” command (sounds counter-intuitive because it is listed as “disconnected” anyway). My question is, will the VMs continue to run and function while I disconnect and reconnect the host back into VC. Thanks.
yeah they will…