OceanStore 2200 V3 - long time switching between controllers while fault

Created: Feb 19, 2020 12:38:18Latest reply: Feb 25, 2020 21:51:17 235 7 0 0
  Rewarded HiCoins: 1 (problem resolved)

Hello,


I got small problem with switching traffic between controller while fault. My infrastructure


Controller A:

ETH1(BOND1)  <-> SW ETH21 (LAG1)

ETH2(BOND1)  <-> SW ETH22 (LAG1)


Controller B:

ETH1(BOND2)  <-> SW ETH23 (LAG2)

ETH2(BOND2)  <-> SW ETH24 (LAG2)


So it looks like ok and it works. On the server site I am using UltraPath. The problem is when I simulate fault on controller A or B I need to wait ~60sec. for switch traffic on second controller. During this time status on ultrapath is not changing. It looks like ultrapath need some time to find fault on both bonded links and switching traffic.

ultrapath

ultrapath_config

ultrapath_config2

ultrapath_config3


Weirder thing is that when I restore even 1 link the status is changing immediately. Is that look for some mistake in UltraPath configuration?


  • x
  • convention:

Featured Answers

Recommended answer

Root.
Created Feb 19, 2020 13:35:47 Helpful(1) Helpful(1)

1. Check the timeout parameter settings. You can change the timeout period of the iSCSI initiator. If you want to speed up failover, set the timeout period to a smaller value, the failover time depends on the I/O hang time of iSCSI initiators(including the link disconnection time detected by NICs ). The failover time cannot be controlled within a certain period.

2. If the network is set up through switches, check the switch configuration also .

3. If the storage is disconnected (for example, the controller is restarted or the port is powered off), analyze the storage behavior process. This may prolong the link disconnection time detected by the host and further prolong the failover time.
  • x
  • convention:

mochab
mochab Created Feb 19, 2020 15:32:07
I found this
https://support.huawei.com/enterprise/en/doc/EDOC1000150159/7584af75/how-do-i-modify-the-iscsi-initiator-s-driver-timeout-time

So in my scenario (I am using Ultrapath) I need to change MaxRequestHoldTime to 5 and LinkDownTime left default?  
mochab
mochab Reply mochab  Created Feb 19, 2020 15:48:18
I have set up MaxRequestHoldTime to 5, LinkDownTime left on default, and after restart the switch traffic goes after ~20sec. This is not 5second but not 60-70sec like before. Do you know why is that work?  
All Answers
lubna
lubna Created Feb 19, 2020 12:50:27 Helpful(0) Helpful(0)

hello dear
open this i hope you get all information
https://www.ico.de/assets/pdf/huawei-2200-description.pdf
  • x
  • convention:

mochab
mochab Created Feb 19, 2020 13:03:39
Nothing about UltraPath and features.  
i%20am%20student%20%20and%20i%20am%20doing%20BSIT%20from%20international%20Islamic%20university%20in%20islamabad
Root.
Root. Created Feb 19, 2020 13:35:47 Helpful(1) Helpful(1)

1. Check the timeout parameter settings. You can change the timeout period of the iSCSI initiator. If you want to speed up failover, set the timeout period to a smaller value, the failover time depends on the I/O hang time of iSCSI initiators(including the link disconnection time detected by NICs ). The failover time cannot be controlled within a certain period.

2. If the network is set up through switches, check the switch configuration also .

3. If the storage is disconnected (for example, the controller is restarted or the port is powered off), analyze the storage behavior process. This may prolong the link disconnection time detected by the host and further prolong the failover time.
  • x
  • convention:

mochab
mochab Created Feb 19, 2020 15:32:07
I found this
https://support.huawei.com/enterprise/en/doc/EDOC1000150159/7584af75/how-do-i-modify-the-iscsi-initiator-s-driver-timeout-time

So in my scenario (I am using Ultrapath) I need to change MaxRequestHoldTime to 5 and LinkDownTime left default?  
mochab
mochab Reply mochab  Created Feb 19, 2020 15:48:18
I have set up MaxRequestHoldTime to 5, LinkDownTime left on default, and after restart the switch traffic goes after ~20sec. This is not 5second but not 60-70sec like before. Do you know why is that work?  
mochab
mochab Created Feb 24, 2020 23:15:18 Helpful(0) Helpful(0)

Guys any suggestions? I have tried to remove Bonding port and leave only eth1 from controller A and eth1 from controller B but stil the same ~20-25sec switch time.
  • x
  • convention:

Ihteshamraza
Ihteshamraza MVE Created Feb 25, 2020 21:51:17 Helpful(0) Helpful(0)

Did you check the logs when this happens?
  • x
  • convention:

I%20am%20Deployment%20and%20Executive%20Engineer%20in%20Corvit%20Networks%20UAE%20(HALP).%20I%20am%20Huawei%20Storage%20expert.%20HCIE%20Storage%2C%20HCIA(Big%20Data%2C%20Cloud%2C%20Wireless%2C%20R%26S).%20I%20am%20taking%20care%20of%20Core%20and%20ICT%20equipment%20for%20customers.%20I%20am%20also%20taking%20care%20of%20customer%20SLAs.%20I%20was%20involved%20in%20many%20projects%20with%20clients%20like%20Google%20and%20Etisalat.%20We%20are%20providing%20support%20and%20maintenance%20for%20their%20data%20centers%20in%20Middle%20East.

Comment

Reply
You need to log in to reply to the post Login | Register

Notice Notice: To protect the legitimate rights and interests of you, the community, and third parties, do not release content that may bring legal risks to all parties, including but are not limited to the following:
  • Politically sensitive content
  • Content concerning pornography, gambling, and drug abuse
  • Content that may disclose or infringe upon others ' commercial secrets, intellectual properties, including trade marks, copyrights, and patents, and personal privacy
Do not share your account and password with others. All operations performed using your account will be regarded as your own actions and all consequences arising therefrom will be borne by you. For details, see " Privacy."
If the attachment button is not available, update the Adobe Flash Player to the latest version!

My Followers

Login and enjoy all the member benefits

Login and enjoy all the member benefits

Login