ONTAP Hardware

Disk Bypass

saranraj456
11,546 Views

What is disk bypass and in what kind of scenario it will happen?

 

Thanks,

Saran

7 REPLIES 7

BrendonHiggins
11,546 Views

Have you seen this yet?

https://kb.netapp.com/support/index?page=content&id=3012395

Drive bypass events are situations which cause the ESH to bypass a  drive port thus making it inaccessible by the host and isolating the  drive from the loop. There are primarily two different kinds of drive  Bypass Events:

  • Threshold Bypasses - Situations where the ESH detects that a  certain kind of error is occurring for a specific period of time and the  determination is made to bypass the port because of accumulated errors  over time or over the amount of data flowing through the loop.
  • Policy Bypasses - Situations where the ESH detects a critical  problem and bypasses a drive in order to maintain the integrity of the  loop. This might be a "situational" problem which later clears, but once  a drive is bypassed, it will remain bypassed until manually cleared by  removing and reseating the drive, or by power cycling of the shelf.
  • Self Bypasses - The hard drive electronics determine that the  internal circuitry cannot function properly anymore. Thus the drive  itself calls to be removed from the loop.

Hope it helps

Bren

saranraj456
11,546 Views

Brendon,

Thank you.

We have one case in netapp for disk failure(or disk went offline).In that case log ,I came across the word disk bypass.so i used it as disk bypass.Is it right to use?

What is the solution for drive bypass?

saranraj456
11,546 Views

How does an ESH module is connected with disks.?

Regards,

Saran

Darkstar
11,546 Views

It can (and will) also happen if you put a disk that only supports 2gbit FC (for example a 10k RPM, or an older 15k RPM disk) into an ESH4 shelf on a 4gbit loop.

SEAJEPROACT
11,546 Views

I just had a situation where a disk was bypassed. "DISK_FAIL - Bypassed by ESH" "Bypassed due to the drive self bypass".

We didn't get any information when checking FAS> aggr status -f. There were no broken disks. The autosupport did show disks on position 11 and 13 but not 12.

We didn't actively do anything to fix this, approx one hour from failure the disk got inserted into the system by itself.


Not sure if this was a Threshold Bypasses?

Couldn't be the Policy Bypass, because we didn't do anything of the following "...it will remain bypassed until manually cleared by removing and reseating the drive, or by power cycling of the shelf."

JGPSHNTAP
11,546 Views

Yeah, this is old news.. Happens on old ESH modules.  Need to manually fail disk or reseat it

SEAJEPROACT
11,546 Views

Check, yea I think it's v 7.3. We didn't do anything, the disk jumped into the system by itself.

Public