Search String: Display: Description: Sort:

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*Yet\s+Another\s+mkfs\.xfs\s+RAID\s+Question\s*$/: 21 ]

Total 21 documents matching your query.

1. Yet Another mkfs.xfs RAID Question (score: 1)
Author: Christian Guggenberger <Christian.Guggenberger@xxxxxxxxxxxxxxxxxxxxxxxx>
Date: Sat, 1 Feb 2003 14:04:27 +0100
I'm going to set up a new 1.7 TB HW Raid 5 next week. RAID 5 contains 11 disks, Stripe Unit should be (according to the manual) 128k. so my mkfs.xfs options will be sunit=256, swidth=2560 for the da
/archives/xfs/2003-02/msg00001.html (8,014 bytes)

2. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Walt H <waltabbyh@xxxxxxxxxxx>
Date: Sat, 01 Feb 2003 09:07:03 -0800
I'm going to set up a new 1.7 TB HW Raid 5 next week. RAID 5 contains 11 disks, Stripe Unit should be (according to the manual) 128k. so my mkfs.xfs options will be sunit=256, swidth=2560 for the dat
/archives/xfs/2003-02/msg00003.html (9,584 bytes)

3. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: "Martin K. Petersen" <mkp@xxxxxxx>
Date: 01 Feb 2003 16:20:46 -0500
Yup. 128KB log alignment seems a bit of an overkill. Does your controller state which chunk size it uses internally? Most controllers use 4-16KB blocks for RAID5. So try aligning your log to values
/archives/xfs/2003-02/msg00005.html (8,987 bytes)

4. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Stephen Lord <lord@xxxxxxx>
Date: 02 Feb 2003 09:11:55 -0600
4K is usually enough to fix performance issues, but it may be dependent on the underlying raid too. Steve
/archives/xfs/2003-02/msg00008.html (9,221 bytes)

5. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Christian Guggenberger <Christian.Guggenberger@xxxxxxxxxxxxxxxxxxxxxxxx>
Date: Sun, 2 Feb 2003 16:27:11 +0100
I could give it try, anyway... This sunit options for the logsection can be overruled at mount time, could'n't they?? I'll check out the manpage. Christian
/archives/xfs/2003-02/msg00010.html (10,320 bytes)

6. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Stephen Lord <lord@xxxxxxx>
Date: 03 Feb 2003 07:52:09 -0600
I don't think you can override log stripe alignment, just the data stripe alignment. Steve
/archives/xfs/2003-02/msg00017.html (8,707 bytes)

7. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Christian Guggenberger <Christian.Guggenberger@xxxxxxxxxxxxxxxxxxxxxxxx>
Date: Mon, 3 Feb 2003 20:52:46 +0100
okay. Any other problems with log v2, sunit=8 I could run into?? Another (little) Question: My machine has 2 GB Ram, and from tomorrow on, 4.3 TB will be exported via kernel-nfsd. Till now, I mount a
/archives/xfs/2003-02/msg00018.html (8,577 bytes)

8. Yet Another mkfs.xfs RAID Question (score: 1)
Author: Christian Guggenberger <Christian.Guggenberger@xxxxxxxxxxxxxxxxxxxxxxxx>
Date: Sat, 1 Feb 2003 14:04:27 +0100
I'm going to set up a new 1.7 TB HW Raid 5 next week. RAID 5 contains 11 disks, Stripe Unit should be (according to the manual) 128k. so my mkfs.xfs options will be sunit=256, swidth=2560 for the da
/archives/xfs/2003-02/msg00485.html (8,014 bytes)

9. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Walt H <waltabbyh@xxxxxxxxxxx>
Date: Sat, 01 Feb 2003 09:07:03 -0800
I'm going to set up a new 1.7 TB HW Raid 5 next week. RAID 5 contains 11 disks, Stripe Unit should be (according to the manual) 128k. so my mkfs.xfs options will be sunit=256, swidth=2560 for the dat
/archives/xfs/2003-02/msg00487.html (9,584 bytes)

10. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: "Martin K. Petersen" <mkp@xxxxxxx>
Date: 01 Feb 2003 16:20:46 -0500
Yup. 128KB log alignment seems a bit of an overkill. Does your controller state which chunk size it uses internally? Most controllers use 4-16KB blocks for RAID5. So try aligning your log to values
/archives/xfs/2003-02/msg00489.html (8,987 bytes)

11. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Stephen Lord <lord@xxxxxxx>
Date: 02 Feb 2003 09:11:55 -0600
4K is usually enough to fix performance issues, but it may be dependent on the underlying raid too. Steve
/archives/xfs/2003-02/msg00492.html (9,221 bytes)

12. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Christian Guggenberger <Christian.Guggenberger@xxxxxxxxxxxxxxxxxxxxxxxx>
Date: Sun, 2 Feb 2003 16:27:11 +0100
I could give it try, anyway... This sunit options for the logsection can be overruled at mount time, could'n't they?? I'll check out the manpage. Christian
/archives/xfs/2003-02/msg00494.html (10,320 bytes)

13. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Stephen Lord <lord@xxxxxxx>
Date: 03 Feb 2003 07:52:09 -0600
I don't think you can override log stripe alignment, just the data stripe alignment. Steve
/archives/xfs/2003-02/msg00501.html (8,707 bytes)

14. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Christian Guggenberger <Christian.Guggenberger@xxxxxxxxxxxxxxxxxxxxxxxx>
Date: Mon, 3 Feb 2003 20:52:46 +0100
okay. Any other problems with log v2, sunit=8 I could run into?? Another (little) Question: My machine has 2 GB Ram, and from tomorrow on, 4.3 TB will be exported via kernel-nfsd. Till now, I mount a
/archives/xfs/2003-02/msg00502.html (8,577 bytes)

15. Yet Another mkfs.xfs RAID Question (score: 1)
Author: Christian Guggenberger <Christian.Guggenberger@xxxxxxxxxxxxxxxxxxxxxxxx>
Date: Sat, 1 Feb 2003 14:04:27 +0100
Hi, I'm going to set up a new 1.7 TB HW Raid 5 next week. RAID 5 contains 11 disks, Stripe Unit should be (according to the manual) 128k. so my mkfs.xfs options will be sunit=256, swidth=2560 for the
/archives/xfs/2003-02/msg00969.html (8,014 bytes)

16. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Walt H <waltabbyh@xxxxxxxxxxx>
Date: Sat, 01 Feb 2003 09:07:03 -0800
I'm going to set up a new 1.7 TB HW Raid 5 next week. RAID 5 contains 11 disks, Stripe Unit should be (according to the manual) 128k. so my mkfs.xfs options will be sunit=256, swidth=2560 for the dat
/archives/xfs/2003-02/msg00971.html (9,600 bytes)

17. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: "Martin K. Petersen" <mkp@xxxxxxx>
Date: 01 Feb 2003 16:20:46 -0500
Yup. 128KB log alignment seems a bit of an overkill. Does your controller state which chunk size it uses internally? Most controllers use 4-16KB blocks for RAID5. So try aligning your log to values
/archives/xfs/2003-02/msg00973.html (9,060 bytes)

18. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Stephen Lord <lord@xxxxxxx>
Date: 02 Feb 2003 09:11:55 -0600
4K is usually enough to fix performance issues, but it may be dependent on the underlying raid too. Steve
/archives/xfs/2003-02/msg00976.html (9,296 bytes)

19. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Christian Guggenberger <Christian.Guggenberger@xxxxxxxxxxxxxxxxxxxxxxxx>
Date: Sun, 2 Feb 2003 16:27:11 +0100
I could give it try, anyway... This sunit options for the logsection can be overruled at mount time, could'n't they?? I'll check out the manpage. Christian
/archives/xfs/2003-02/msg00978.html (10,434 bytes)

20. Re: Yet Another mkfs.xfs RAID Question (score: 1)
Author: Stephen Lord <lord@xxxxxxx>
Date: 03 Feb 2003 07:52:09 -0600
I don't think you can override log stripe alignment, just the data stripe alignment. Steve
/archives/xfs/2003-02/msg00985.html (8,850 bytes)

Current List: 1 - 20
Page: [1] [2]
This search system is powered by Namazu