This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Link Aggregation for clustering/HA

This is a throughput issue as well as an up-time issue.

For the first year of V7, there was no way to have the SMTP proxy blackhole spam, so the clustered 220s at one client site were permanently pegged at 100%+ with the traffic on the 100MB connection between the units. One Astaro engineer opined that the client should have 320s so they could connect at a gigabit. Today, with 14 days in the quarantine instead of three (prior to V7.300), the cluster rarely averages above 35%.

If the HA link fails, a cluster stops updating the slave, and the slave no longer knows when to take over.

This thread was automatically locked due to age.

0 crashtestdummy over 17 years ago

That sounds similar to my experiences. They've made some huge improvements to mail throughput in 7.3.

If you disable/re-enable SMTP after the failover, does it start mail flowing again?
Cancel
Vote Up 0 Vote Down

Cancel
0 BAlfson over 17 years ago in reply to crashtestdummy

We never tried that. Just opted for rebooting the master then the slave
Cancel
Vote Up 0 Vote Down

Cancel
0 BAlfson over 17 years ago in reply to BAlfson

I discovered yesterday that the 'Backup interface' box, which lets you select the internal or external interface, does provide the function I requested above.

Change the request to: "On the Configuration tab, please add an explanation of the function of the backup interface. On the otherhand, if the box were labeled 'Backup connection via', it would be clearer."
Cancel
Vote Up 0 Vote Down

Cancel