JulianMilano
2 Bronze

How do Consistency Group (CG) bandwith limitations add up?

Jump to solution

If I have 4 CGs, and they are respectively set to 5, 10, 15 & 20Mbps bandwidth limitation via the CLI, would I expect to see usage at up to 50Mbps on initial sync?

The reason I ask is that my colleague setup 4 CGs to replicate about a month ago and it was working OK until last week when our WAN usage went 99% according to Orion monitoring tools. We eventually found that it was the RPAs, and to stop it we turned off the replication of the CGs and the issue was fixed, but we cannot figger out what happened to cause the bandwidth usage to go so high.

The only thing I can think of is that the CGs got so far out of sync, or so much data changed in the CGs that the amount of replication was huge, but I don't know how to find that out.

PS. We don't have compression or dedupe setup yet and we are running RecoverPoint 4.0. The WAN link is 50Mbps max.

Labels (1)
Tags (1)
0 Kudos
1 Solution

Accepted Solutions
forshr
4 Tellurium

Re: How do Consistency Group (CG) bandwith limitations add up?

Jump to solution

The feature is set per RPA via the config_link_policy CLI command.

0 Kudos
4 Replies
JulianMilano
2 Bronze

Re: How do Consistency Group (CG) bandwith limitations add up?

Jump to solution

I spoke to a colleague of mine about this who stated the following:

I found out Consistency Group SERVER_DRIVE_BACKUP_700GB_PROD has caused the bandwidth flood.

It was residing on RPA2 on Sydney RPA Cluster , it was set to 10Mbps speed, however, due some reason after it replicated for a week, this group has switched to RPA1 , which I believe it was not set to 10Mbps, the default value is “Unlimited”.

So lesson learnt , we need to set both RPA1 and RPA2 on all the CG, no matter it resides on which RPA.

But I don't understand how this could be- aren't the RPAs in a cluster so why would we have to set properties on both nodes of an RPA cluster?

0 Kudos
forshr
4 Tellurium

Re: How do Consistency Group (CG) bandwith limitations add up?

Jump to solution

The feature works as follows:

The system sums bandwidth limitations of all consistency groups on a single RPA and limits the outgoing throughput from the RPA to that sum. If consistency groups with settings of ‘limited’ and ‘unlimited’ run on the same RPA, the effect is that there is no limit to the outgoing throughput from the RPA.

Regards,

Rich Forshaw

RecoverPoint Corporate Systems Engineering

0 Kudos
Highlighted
JulianMilano
2 Bronze

Re: How do Consistency Group (CG) bandwith limitations add up?

Jump to solution

Thanks Rich, so when setting the bandwidth values, do I have to set it on both RPAs in the same cluster or do I set it once and it replicates the values to both RPAs?

0 Kudos
forshr
4 Tellurium

Re: How do Consistency Group (CG) bandwith limitations add up?

Jump to solution

The feature is set per RPA via the config_link_policy CLI command.

0 Kudos