Operating system version = Linux Redhat 6.2
Storage Foundation version = 6.0 with RP1
The below scerario as a reference:
See the below errors and details which I am watching under dmesg command of linux. furthermore occasionally the message "host become <unreachable>" is being come with repstatus command.
The details below has been drive from the below mentioned link.
Disconnecting rlink rlink_name due to stream error
VVR disconnected the named RLINK for the reason mentioned in the message.
Action: Ignore message if seen occasionally. VVR automatically recovers from the error.
If the errors persist, they may indicate a problem with the network configuration.
Rlink rlink_name disconnecting due to ack timeout on msg type message
Ignore message if seen occasionally.
Action: If the message persists, contact Veritas Customer Support.
The connectivity between primary and secondary is suppose 25Mbps and the primary and secondary host is sharing this link. Moreover many other OS/machines also sharing this same link. From ISP there is a stretch of bandwidth utilization up to 30Mbps. After reaching above 30Mbps the bandwidth link gets chook and the bandwidth drops from 30Mbps to 0Mbps and after a while the link again start getting the bandwidth, increasing from 0Mbps. but that thing happening again and again as the data increases the bandwidth again reach above 30Mbps and again gets chook. We feel that this happens when we copy the data(around few GB's) at primary site just because to check the replication behaviour OR either enough data is being written at primary site.
We just set the bandwidth limit with the vxedit command via using the option of bandwidth_limit. This restrict the replication traffic to use just that bandwidth which is allowed.
Please Note: There may be some other culprits which may generate the above mentioned errors and the circumstances we I faced but the resolution I shared is worked for me. I hope this may help.
Helpful Link for more details
For Bandwidth Control:
For Bandwidth usage details: