We are in the process of migrating this forum. A new space will be available soon. We are sorry for the inconvenience.

RPS Packet loss/latency at certain time each day


jameslewis
17-04-2010, 15:25
RPS network in failure again today, terrible packetloss, download speed of about 3KB/s....!!

Edit:
Back to normal again now...

marks
14-04-2010, 09:25
For rapid check of the issue as soon as it happens, make sure that the OVH SSH key is installed.

For this 2 RPS, please give as the rps name, IP or ticket number so we can check if there is work or problem detected on them.

If you want, we can also deal directly with you on the support. Send us an email with the details.

Edit: keep an eye on the task http://travaux.ovh.net/?do=details&id=4097. Your RPS is affected by this task. Our engineers are working on it and should be fixed soon.

jameslewis
13-04-2010, 16:37
It was OK earlier today but now it has gone crazy again on both RPS this time, major packet loss.

Neil
13-04-2010, 09:59
Quote Originally Posted by jameslewis
I have had 2 RPS servers for about 2 years now.

Over the past few weeks, one of the RPS servers experiences a lot of packet loss for periods in the afternoon/evening from around 16:00 BST/UK each day.

The other RPS works fine during this period. The packet loss & poor connectivity comes and goes and lasts for a few hours each afternoon/evening. It is bad from many locations using http://just-ping.com/ to test.

I've not changed anything on the servers, connectivity to my other RPS is fine at the same period that the other is experiencing bad packetloss. They are both set up with the same OS and software also.

I have raised ticket 413748 some days ago which was closed by OVH as by the time it was checked the packetloss was gone. I reopened the ticket as the problem comes back in the afternoon each day. I reopened the ticket again, still waiting for some resolution. I have also sent pingplotter charts showing how bad it gets.

Looking at MRTG graph at the time the traffic drops right off when it happens (obviously?!)

I am wondering if something in the network near the problem RPS is causing the issue or is getting overloaded? Problem server is in Bay : 08F01
(server with no issue is Bay : 08G05 )
OVH manager is showing both servers as 100mbs unlimited.

Any ideas?


Ping statistics for 87.98.134.xx:
Packets: Sent = 50, Received = 38, Lost = 12 (24%
loss),
Approximate round trip times in milli-seconds:
Minimum = 19ms, Maximum = 43ms, Average = 24ms



Ping statistics for 87.98.134.xx:
Packets: Sent = 20, Received = 13, Lost = 7 (35%
loss),
Approximate round trip times in milli-seconds:
Minimum = 42ms, Maximum = 95ms, Average = 54ms


> ethtool eth0
Settings for eth0:
Supported ports: [ TP MII ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
Supports auto-negotiation: Yes
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
Advertised auto-negotiation: Yes
Speed: 100Mb/s
Duplex: Full
Port: MII
PHYAD: 1
Transceiver: internal
Auto-negotiation: on
Supports Wake-on: pg
Wake-on: d
Current message level: 0x000000c5 (197)
Link detected: yes



CPU load averages 0.03 (1 min) 0.20 (5 mins) 0.12 (15 mins)
Real memory 454.81 MB total, 102.68 MB used

Virtual memory 491.99 MB total, 0 bytes used

Local disk space 10.22 GB total, 971.86 MB used

The incident team asked for the SSH Key to be installed, please can you do this.

Quote Originally Posted by chazor
It doesnt only happen on RPS.Hybrid range got 60% ping lost of 24 HOURS to some euro countries in past 2 days,OVH dont bother to fix it yet. However my EG MAX has no problem at all.

Ping statistics for 83.149.80.111: (leaseweb.com)
Packets: Sent = 266, Received = 108, Lost = 158 (59% loss),
Chazor server address, it may of been affected by http://translate.google.com/translat...&sl=auto&tl=en

chazor
13-04-2010, 02:02
It doesnt only happen on RPS.Hybrid range got 60% ping lost of 24 HOURS to some euro countries in past 2 days,OVH dont bother to fix it yet. However my EG MAX has no problem at all.

Ping statistics for 83.149.80.111: (leaseweb.com)
Packets: Sent = 266, Received = 108, Lost = 158 (59% loss),

jameslewis
12-04-2010, 20:27
I have had 2 RPS servers for about 2 years now.

Over the past few weeks, one of the RPS servers experiences a lot of packet loss for periods in the afternoon/evening from around 16:00 BST/UK each day.

The other RPS works fine during this period. The packet loss & poor connectivity comes and goes and lasts for a few hours each afternoon/evening. It is bad from many locations using http://just-ping.com/ to test.

I've not changed anything on the servers, connectivity to my other RPS is fine at the same period that the other is experiencing bad packetloss. They are both set up with the same OS and software also.

I have raised ticket 413748 some days ago which was closed by OVH as by the time it was checked the packetloss was gone. I reopened the ticket as the problem comes back in the afternoon each day. I reopened the ticket again, still waiting for some resolution. I have also sent pingplotter charts showing how bad it gets.

Looking at MRTG graph at the time the traffic drops right off when it happens (obviously?!)

I am wondering if something in the network near the problem RPS is causing the issue or is getting overloaded? Problem server is in Bay : 08F01
(server with no issue is Bay : 08G05 )
OVH manager is showing both servers as 100mbs unlimited.

Any ideas?


Ping statistics for 87.98.134.xx:
Packets: Sent = 50, Received = 38, Lost = 12 (24%
loss),
Approximate round trip times in milli-seconds:
Minimum = 19ms, Maximum = 43ms, Average = 24ms



Ping statistics for 87.98.134.xx:
Packets: Sent = 20, Received = 13, Lost = 7 (35%
loss),
Approximate round trip times in milli-seconds:
Minimum = 42ms, Maximum = 95ms, Average = 54ms


> ethtool eth0
Settings for eth0:
Supported ports: [ TP MII ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
Supports auto-negotiation: Yes
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
Advertised auto-negotiation: Yes
Speed: 100Mb/s
Duplex: Full
Port: MII
PHYAD: 1
Transceiver: internal
Auto-negotiation: on
Supports Wake-on: pg
Wake-on: d
Current message level: 0x000000c5 (197)
Link detected: yes



CPU load averages 0.03 (1 min) 0.20 (5 mins) 0.12 (15 mins)
Real memory 454.81 MB total, 102.68 MB used

Virtual memory 491.99 MB total, 0 bytes used

Local disk space 10.22 GB total, 971.86 MB used