Services are interrupted for several seconds during a stack switchover.

Created: May 28, 2019 21:51:50Latest reply: May 28, 2019 22:03:00 75 3 0 0
  Rewarded Hi-coins: 0 (problem resolved)

214930dl6kr4gf2lbrrlg2.png?stack.png

As shown in the preceding figure, at each site, two S5720 switches are stacked using stack cards. The stacked switches are connected to transmission devices through Eth-Trunk interfaces on different cards. OSPF is enabled on the uplink interface, and the direct route is redistributed. The link aggregation mode is manual.

Normal communication is normal. However, when the master member of the stack switch is faulty (powered off), service communication is interrupted for about 10 seconds. In addition, when the old master is restarted, services are interrupted for about 30 seconds. Is this normal? Is there a way to reduce the stack failover delay?

Services are interrupted for several seconds during a stack switchover.

  • x
  • convention:

Featured Answers
Popeye_Wang  Enthusiast Technician   Created May 28, 2019 22:03:00 Helpful(0) Helpful(0)

Posted by Hobbit at 2019-05-28 22:00After STP is disabled, the communication interruption problem is resolved after the faulty device  ...
The method of calculating the STP cost value of an Eth-trunk interface is the cost value of a single member interface / the number of member interfaces. 

If the STP cost value of an Eth-Trunk member interface up/down changes, STP recalculation is triggered and the interface reconverges. During the convergence, downstream link server’s interface wait for two times forwarding time (30 seconds), traffic is interrupted.

In addition, the system MAC address switching may also cause STP flapping. Generally, MAC address of the master switch is used as the stack MAC address. When the master switch changes, the stack system MAC address changes accordingly. By default, the system MAC address switching delay is 10 minutes. Check whether the immediate switchover command is configured: undo stack timer mac-address switch-delay.


And modifying the Hello intervals is not useful. You can try configuring OSPF GR.

OSPF GR:

Example for Configuring OSPF GR:
  • x
  • convention:

All Answers
Popeye_Wang  Enthusiast Technician   Created May 28, 2019 21:52:59 Helpful(0) Helpful(0)

The 10s is normal, it’s should be the time when the OSPF neighbor relationship is re-established and the route is refreshed. The OSPF neighbor relationship is established with the system master. If the active node is powered off, the OSPF neighbor relationship is re-established.
The second 30s may be caused by STP convergence. You can check whether the STP port status changes during the switchover. If yes, disable STP in system view or enable STP edge-port on the port connected to the server.
  • x
  • convention:

Hobbit     Created May 28, 2019 22:00:20 Helpful(0) Helpful(0)

Posted by Popeye_Wang at 2019-05-28 21:52 The 10s is normal, it’s should be the time when the OSPF neighbor relationship is re-established an ...
After STP is disabled, the communication interruption problem is resolved after the faulty device is powered on. Why does stp affect the switchover when no Loop Occurs?

And the recovery time of the power-off switchover is still 10 seconds. If the 10s time-consuming is mainly caused by OSPF reset neighbors, what other methods can optimize this time besides modifying the sending interval of OSPF Hello messages?
  • x
  • convention:

Popeye_Wang  Enthusiast Technician   Created May 28, 2019 22:03:00 Helpful(0) Helpful(0)

Posted by Hobbit at 2019-05-28 22:00After STP is disabled, the communication interruption problem is resolved after the faulty device  ...
The method of calculating the STP cost value of an Eth-trunk interface is the cost value of a single member interface / the number of member interfaces. 

If the STP cost value of an Eth-Trunk member interface up/down changes, STP recalculation is triggered and the interface reconverges. During the convergence, downstream link server’s interface wait for two times forwarding time (30 seconds), traffic is interrupted.

In addition, the system MAC address switching may also cause STP flapping. Generally, MAC address of the master switch is used as the stack MAC address. When the master switch changes, the stack system MAC address changes accordingly. By default, the system MAC address switching delay is 10 minutes. Check whether the immediate switchover command is configured: undo stack timer mac-address switch-delay.


And modifying the Hello intervals is not useful. You can try configuring OSPF GR.

OSPF GR:

Example for Configuring OSPF GR:
  • x
  • convention:

Reply

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

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!

Login and enjoy all the member benefits

Login
Fast reply Scroll to top