Tuesday, February 07, 2012 9:50 PM
I have a Plesk CentOS 5.5 VM running in Hyper-V that seems to drop about 1% of its packets. The host itself does not drop any packets, just the VM. I've tested having both the host (Server 2008) and the client (CentOS 5.5) simultaneously and I can see that the host doesn't have any interuptions while the client does.
seth0 Link encap:Ethernet HWaddr 00:15:5D:01:6F:00
inet addr:x.x.x.x Bcast:x.x.x.x Mask:255.255.255.248
inet6 addr: fe80::215:5dff:fe01:6f00/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:209304597 errors:0 dropped:0 overruns:0 frame:0
TX packets:194666860 errors:0 dropped:0 overruns:0 carrier:0
RX bytes:120617262583 (112.3 GiB) TX bytes:108309221137 (100.8 GiB)
2090 packets transmitted, 2061 received, 1% packet loss, time 2108778ms
rtt min/avg/max/mdev = 0.000/7.710/980.522/41.922 ms
While the adapter status shows everything is fine (whatever it is checking); ping, ssh, and snmp drop continually. Is there anything that needs to be done specifically for a Redhat NIC to work properly inside Hyper-V?
- Edited by lekcomp Tuesday, February 07, 2012 9:54 PM added detail
Saturday, February 11, 2012 11:36 PM
Before we continue this, you have installed Linux Integration services in the guest OS? from here:
they will smooth out some odd issues you might experience =)
- Marked As Answer by Vincent HuModerator Monday, February 13, 2012 9:10 AM
Thursday, February 16, 2012 2:26 PMI already installed the v2.1 about a year ago. Does it need to be upgraded? The pdf for that version shows it supports: Red Hat Enterprise Linux 6.0 and 6.1 x86 and x64. No mention of previous versions.
- Edited by lekcomp Thursday, February 16, 2012 5:18 PM Additional Info
Tuesday, April 10, 2012 7:17 PMWe encounter the same problems. Random packet drops running CentOS 5.8 under Hyper-V R2 (fully updated). Do you really recommend upgrading to Linux Integration Services >= 3 with CentOS 5?