github-starred/zfs-ha
2
0
Fork 0
mirror of https://github.com/ewwhite/zfs-ha.git synced 2026-05-15 22:05:04 -06:00
Code Issues 11 Projects Packages Wiki Activity Actions

[GH-ISSUE #39] Why not use iSCSI instead of SAS? #38

New issue
Closed
opened 2026-05-05 03:32:29 -06:00 by gitea-mirror · 1 comment
gitea-mirror commented 2026-05-05 03:32:29 -06:00
Owner
Copy link

Originally created by @snaix on GitHub (Jun 7, 2022).
Original GitHub issue: https://github.com/ewwhite/zfs-ha/issues/39

Hi @ewwhite

The zfs-ha is cool. After reading the wiki, I have an idea about replacing SAS cable with ethernet cable, and use a iSCSI disk over LAN.

look at this image:
image

If just replace SAS cable with ethernet cable, it seems everything is fine, they have same logic: just replace a cable type, nothing changed on ZFS.
Of course, with iSCSI, there are more coupling, like NIC, Cable, Switch, and TCP congestion problem.
But blocks transmission over Ethernet provides more flexibility: we can use any server to be a ZFS head node, when the old one was crashed. And also, we can use an independent switch (and iscsi multi-path) for storage to avoid switch SPOF, congestion problem...

The performance problem doesn't seem serious at present: 10/25/100GE is more and more popular, faster than SAS3 12G.
And iSER...

So is there any problem with this idea?

Originally created by @snaix on GitHub (Jun 7, 2022). Original GitHub issue: https://github.com/ewwhite/zfs-ha/issues/39 Hi @ewwhite The zfs-ha is cool. After reading the wiki, I have an idea about replacing SAS cable with ethernet cable, and use a iSCSI disk over LAN. look at this image: ![image](https://user-images.githubusercontent.com/6660768/172614018-4da9a3be-902b-4d02-9688-ee99c520a63f.png) If just replace SAS cable with ethernet cable, it seems everything is fine, they have same logic: just replace a cable type, nothing changed on ZFS. Of course, with iSCSI, there are more coupling, like NIC, Cable, Switch, and TCP congestion problem. But blocks transmission over Ethernet provides more flexibility: we can use any server to be a ZFS head node, when the old one was crashed. And also, we can use an independent switch (and iscsi multi-path) for storage to avoid switch SPOF, congestion problem... The performance problem doesn't seem serious at present: 10/25/100GE is more and more popular, faster than SAS3 12G. And iSER... So is there any problem with this idea?
gitea-mirror commented 2026-05-05 03:32:36 -06:00
Author
Owner
Copy link

@ewwhite commented on GitHub (Aug 8, 2022):

Hello,

This is definitely something you can explore.
It's out of scope for this project, though.

<!-- gh-comment-id:1208139969 --> @ewwhite commented on GitHub (Aug 8, 2022): Hello, This is definitely something you can explore. It's out of scope for this project, though.
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
master
No results found.
Labels
Clear labels   
pull-request

Mirrored from GitHub Pull Request

No labels
pull-request
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/zfs-ha#38
No description provided.
Delete branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?