Unsolved
This post is more than 5 years old
79 Posts
0
1577
Removing an assumed unused LUN from RecoverPoint - Fibre Channel packets lost
Hi,
I inherited a RecoverPoint configuration a while back, and for the most part I'm fairly comfortable managing it.
Recently, we virtualized a host with some LUNs from a VNX. This host did not have its LUNs replicated or in any other way managed by RecoverPoint, but I saw that one of the LUNs was presented to an RPA.
I have no idea why this was the case, perhaps the host originally had replicated data.
Nowhere in RecoverPoint, or at least in the management GUI, can I find any reference to this LUN. Nowhere in any replication set or journal LUN.
I unpresented it from RecoverPoint, and after about a minute I get 15-20 warnings: "10 Fibre Channel packets were lost during the past 60 minutes from hba_0 to 0x5006016946e03f72. Possible causes: slow storage, bad Fibre Channel cable, or other connectivity issue. ;"
No consistency groups get any warning or error, and I have no complaints from any of the running hosts.
Is this just a warning saying RP lost connectivity to a LUN, but RP wasn't using it for anything? Can I clean this up somehow?
forshr
1.1K Posts
1
September 23rd, 2014 05:00
There are multiple reasons for this type of message. In your circumstances I would consider doing the following as it may be related to an inquiry command from an RPA based on residual visibility of the volume that you removed from the RP SG.
1. Run the rescan_san CLI command and wait to see if the message repeats itself which is on an 60 minute cyclical basis.
2. If the mesaage returns, log onto each RPA in the cluster as boxmgmt and run the SAN Diagnostics command (called Run SAN diagonstics).
3. If the message returns, reboot each RPA in the cluster in turn.
4. If the message returns open an SR.
Regards,
Rich
sechm
79 Posts
0
September 23rd, 2014 07:00
After an hour or so the warnings went away. Probably just as I thought, just a notification a LUN was no longer accessible.