whitetiger264 commented on issue #6836:
URL: https://github.com/apache/cloudstack/issues/6836#issuecomment-1291952986

   Hi Guys
   
   Just another update. I added 1 Gbps port speeds at both my KVM node and 
MGMNT node where the storage drives are. This has help significantly lower the 
time execution limit that was previously set to 3600 seconds. I can now work 
with this limit. However, should VMs increase in  size I will run back into 
this error.
   
   The problem that I now experience is that the transfer speeds are limited to 
100 Mbps, but I have 1 Gbps in place. I spoke with the ISP and they said the 
following:
   
   > Our network engineers have advised that "both interfaces for xxxxx and 
xxxxxx are configured for 1Gbs operation. We do not rate-limit client traffic 
in the data-centre except through port-based general bandwidth limiters.
   >
   > The client must do iperf tests in both directions. To do a reverse 
direction test, add the '-R' parameter to the client side command.
   >
   > If the iperf tests indicate 1Gbs in both directions, the client should 
examine the copy process more closely. 
   > What protocol is involved? 
   > Is there some inherent speed limit to the protocol? 
   > Is there parameters related to the copy daemon or client which could 
influence the through-put."
   
   Could you kindly confirm what protocols are involved during the volume 
snapshot process? Are there any speed limits in the global configurations that 
we need to look at? Any feedback on this matter will be appreciated. 
   
   Here is an `iperf` test between the two servers:
   
   `iperf server`:
   
   ```
   root@hostname ~ $ iperf -s -i 10
   ------------------------------------------------------------
   Server listening on TCP port 5001
   TCP window size:  256 KByte (default)
   ------------------------------------------------------------
   [  1] local KVMHOSTIP port 5001 connected with MGMNTHOSTIP port 57880 
(reverse)
   [ ID] Interval       Transfer     Bandwidth
   [  1] 0.00-10.00 sec  1.09 GBytes   934 Mbits/sec
   [  1] 0.00-10.02 sec  1.09 GBytes   932 Mbits/sec
   [  2] local KVMHOSTIP port 5001 connected with MGMNTHOSTIP port 48754 
(reverse)
   [ ID] Interval       Transfer     Bandwidth
   [  2] 0.00-10.00 sec  1.07 GBytes   923 Mbits/sec
   [  2] 0.00-10.03 sec  1.08 GBytes   921 Mbits/sec
   ```
   
   
   `iperf client`:
   
   ```
   root@hostname ~ $ iperf -i 10 -c KVMHOSTIP -R
   ------------------------------------------------------------
   Client connecting to KVMHOSTIP, TCP port 5001
   TCP window size:  128 KByte (default)
   ------------------------------------------------------------
   [  1] local MGMNTHOSTIP port 48754 connected with KVMHOSTIP port 5001 
(reverse)
   [ ID] Interval       Transfer     Bandwidth
   [ *1] 0.00-10.00 sec  1.07 GBytes   923 Mbits/sec
   [ *1] 0.00-10.01 sec  1.08 GBytes   923 Mbits/sec
   root@hostname ~ $ 
   
   ```
   
   But here you can see, during the volume snapshot process, `iftop` shows that 
speeds remain close to 100 Mbps as if there's a limit in place somewhere:
   
   ![Screenshot from 2022-10-25 
11-04-39](https://user-images.githubusercontent.com/16236376/198024985-41958cdc-f46b-465d-a38e-a59cd0917112.png)


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to