[fixed] /var on separate xfs partition

As for topic, I recently moved my /var directory to a separate partition, formatted as xfs, mounted in /mnt. Problem is: after that I experienced slowdowns when using pacman (with high disk activity). I think the problem might be xfs (as I've read pacman is slow on that) or maybe pacman has problem using the link in /var to my mounted partition?
I'd like to know whether there's some obvious problem before trying to change filesystem type (which one, btw?)...
edit: fixed by changing fs type
Last edited by sunriis (2007-09-22 15:30:40)

Hi,
Maybe it would be better to let the logfiles in /private/var/log untouched. You can rotate the logfiles to another disk, without having trouble. See /etc/weekly (daily / monthly) there you can edit the path.
Safer against updates is to create your own files in /etc/yourconfigfiles/ . I use this for rotating my logfiles, because logfiles like secure.log, system.log and asr.log grows very fast and I don't want xGigabyte big logfiles, which are often opened from my system.
Don't touch /etc/fstab , because it is deprecated and maybe this will disappear in next versions.
Write you rotatescripts:
/etc/yourconfigfiles/
daily.my
weekly.my
monthly.my
Edit rotate files:
/etc/daily
/etc/weekly
/etc/monthly
Put to them a line like following:
. /etc/yourconfigfiles/daily.my
if you want to exit manually before the system script does unimportant things, like rotating your rotated logfiles then do an exit after your . /etc/yourconfigfiles/* entry.
hope this helps,
sandor szücs
Mac OS X (10.4.6)

Similar Messages

[Solved]Grow an XFS partition

I have /dev/system/home which has been enlarged by 20GB to 60GB, the XFS partition on it is 40GB in size. When I view the help section I don't know what option to use:
[root@JORDAN-CD3CDA3B jordanwb]# xfs_growfs --help
Usage: xfs_growfs [options] mountpoint
Options:
-d grow data/metadata section
-l grow log section
-r grow realtime section
-n don't change anything, just show geometry
-I allow inode numbers to exceed 32 significant bits
-i convert log from external to internal format
-t alternate location for mount table (/etc/mtab)
-x convert log from internal to external format
-D size grow data/metadata section to size blks
-L size grow/shrink log section to size blks
-R size grow realtime section to size blks
-e size set realtime extent size to size blks
-m imaxpct set inode max percent to imaxpct
-V print version information
Last edited by jordanwb (2008-10-13 00:33:49)

According to this site:
http://www.linux.sgi.com/archives/xfs/2 … 00008.html
You just run `xfs_growfs /dev/system/home` to let it fill the available space on the device

Mounting separate /home partition

So I have a separate /home partition.
It's /dev/sda1 and a EXT3 partition.
How would I mount this as /home in the /etc/fstab?

sand_man wrote:
Arm-the-Homeless wrote:
So I have a separate /home partition.
It's /dev/sda1 and a EXT3 partition.
How would I mount this as /home in the /etc/fstab?
I understand that this is the Newbie Corner but this should have already been done in the Arch Setup. Or are you doing a manual install?
I already had a separate /home partition and when I selected it, it said to pick a filesystem.
So I thought that meant it would be overwritten.

[SOLVED] XFS partition or Harddrive failing

Hello
My /home XFS-formatted partition is in different disk (seagate 750G). Lately it has begun failing constantly when doing lots of copying/moving work with following type of errors:
- cannot read first 512 bytes
- mount: special device "dev/mount-by-label/home" does not exist
The last one happened during normal boot. After powering off, waiting a while and booting again all was good - until I tried to backup my data, when midway while copying I got:
FXS metadata write error block 0x403e880 in sdb1
I/O error in filesystem meta-data ("xlog_iodone") error 5 buf count 2560
Filesystem: Log I/O Error Detected
I tried to find solution from several forums, but all I got was to try "xfs_repair". It doesn't find any errors when the partition/drive is working properly, but after I tried backup my data and got those errors above, "xfs_repair" gives following:
[root@golved ~]# xfs_repair /dev/sda1
Phase 1 - find and verify superblock...
superblock read failed, offset 0, size 524288, ag 0, rval -1
fatal error -- Input/Output error
I am currently trying "smartmontools", but at leas now it doesn't work. Maybe after reboot it gives me some sensible output. Meanwhile, can any of you give me any pointers how to find out is the problem in XFS filesystem or in my year old Seagate Barracuda 7200.11/12(?) 750G harddrive? How about getting my data out, is there way to bypass the logging that xfs does? I am thinking about that because of those "Log I/O errors"... Currently, whenever the drive mounts properly, I can do some copying but it halts eventually. Maybe disabling logging and all other non-needed things from xfs would let me backup more data before I/O errors arise?
Last edited by mclang (2009-03-11 06:43:21)

Okay, after couple of boots smartctl gave following:
smartctl version 5.38 [i686-pc-linux-gnu] Copyright (C) 2002-8 Bruce Allen
Home page is http://smartmontools.sourceforge.net/
=== START OF INFORMATION SECTION ===
Model Family: Seagate Barracuda 7200.11
Device Model: ST3750330AS
Serial Number: 9QK06CMQ
Firmware Version: SD15
User Capacity: 750 156 374 016 bytes
Device is: In smartctl database [for details use: -P show]
ATA Version is: 8
ATA Standard is: ATA-8-ACS revision 4
Local Time is: Sun Mar 8 18:42:27 2009 EET
SMART support is: Available - device has SMART capability.
SMART support is: Enabled
=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED
General SMART Values:
Offline data collection status: (0x82) Offline data collection activity
was completed without error.
Auto Offline Data Collection: Enabled.
Self-test execution status: ( 121) The previous self-test completed having
the read element of the test failed.
Total time to complete Offline
data collection: ( 634) seconds.
Offline data collection
capabilities: (0x7b) SMART execute Offline immediate.
Auto Offline data collection on/off support.
Suspend Offline collection upon new
command.
Offline surface scan supported.
Self-test supported.
Conveyance Self-test supported.
Selective Self-test supported.
SMART capabilities: (0x0003) Saves SMART data before entering
power-saving mode.
Supports SMART auto save timer.
Error logging capability: (0x01) Error logging supported.
General Purpose Logging supported.
Short self-test routine
recommended polling time: ( 1) minutes.
Extended self-test routine
recommended polling time: ( 172) minutes.
Conveyance self-test routine
recommended polling time: ( 2) minutes.
SCT capabilities: (0x103b) SCT Status supported.
SCT Feature Control supported.
SCT Data Table supported.
SMART Attributes Data Structure revision number: 10
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE
1 Raw_Read_Error_Rate 0x000f 101 097 006 Pre-fail Always - 52655725
3 Spin_Up_Time 0x0003 093 093 000 Pre-fail Always - 0
4 Start_Stop_Count 0x0032 100 100 020 Old_age Always - 26
5 Reallocated_Sector_Ct 0x0033 100 100 036 Pre-fail Always - 2043
7 Seek_Error_Rate 0x000f 070 060 030 Pre-fail Always - 8613925167
9 Power_On_Hours 0x0032 091 091 000 Old_age Always - 8431
10 Spin_Retry_Count 0x0013 100 100 097 Pre-fail Always - 0
12 Power_Cycle_Count 0x0032 100 100 020 Old_age Always - 26
184 Unknown_Attribute 0x0032 100 100 099 Old_age Always - 0
187 Reported_Uncorrect 0x0032 031 031 000 Old_age Always - 69
188 Unknown_Attribute 0x0032 100 099 000 Old_age Always - 8590065666
189 High_Fly_Writes 0x003a 100 100 000 Old_age Always - 0
190 Airflow_Temperature_Cel 0x0022 070 055 045 Old_age Always - 30 (Lifetime Min/Max 28/32)
194 Temperature_Celsius 0x0022 030 045 000 Old_age Always - 30 (0 21 0 0)
195 Hardware_ECC_Recovered 0x001a 047 022 000 Old_age Always - 52655725
197 Current_Pending_Sector 0x0012 100 100 000 Old_age Always - 4
198 Offline_Uncorrectable 0x0010 100 100 000 Old_age Offline - 4
199 UDMA_CRC_Error_Count 0x003e 200 200 000 Old_age Always - 0
SMART Error Log Version: 1
ATA Error Count: 100 (device log contains only the most recent five errors)
CR = Command Register [HEX]
FR = Features Register [HEX]
SC = Sector Count Register [HEX]
SN = Sector Number Register [HEX]
CL = Cylinder Low Register [HEX]
CH = Cylinder High Register [HEX]
DH = Device/Head Register [HEX]
DC = Device Command Register [HEX]
ER = Error register [HEX]
ST = Status register [HEX]
Powered_Up_Time is measured from power on, and printed as
DDd+hh:mm:SS.sss where DD=days, hh=hours, mm=minutes,
SS=sec, and sss=millisec. It "wraps" after 49.710 days.
Error 100 occurred at disk power-on lifetime: 8426 hours (351 days + 2 hours)
When the command that caused the error occurred, the device was active or idle.
After command completion occurred, registers were:
ER ST SC SN CL CH DH
04 71 04 9d 00 32 e0 Device Fault; Error: ABRT
Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
a1 00 00 00 00 00 a0 02 00:27:26.893 IDENTIFY PACKET DEVICE
ec 00 00 00 00 00 a0 02 00:27:26.887 IDENTIFY DEVICE
00 00 00 00 00 00 00 04 00:27:26.733 NOP [Abort queued commands]
a1 00 00 00 00 00 a0 02 00:27:21.427 IDENTIFY PACKET DEVICE
ec 00 00 00 00 00 a0 02 00:27:21.420 IDENTIFY DEVICE
Error 99 occurred at disk power-on lifetime: 8426 hours (351 days + 2 hours)
When the command that caused the error occurred, the device was active or idle.
After command completion occurred, registers were:
ER ST SC SN CL CH DH
04 71 04 9d 00 32 e0
Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
ec 00 00 00 00 00 a0 02 00:27:26.887 IDENTIFY DEVICE
00 00 00 00 00 00 00 04 00:27:26.733 NOP [Abort queued commands]
a1 00 00 00 00 00 a0 02 00:27:21.427 IDENTIFY PACKET DEVICE
ec 00 00 00 00 00 a0 02 00:27:21.420 IDENTIFY DEVICE
00 00 00 00 00 00 00 04 00:27:21.267 NOP [Abort queued commands]
Error 98 occurred at disk power-on lifetime: 8426 hours (351 days + 2 hours)
When the command that caused the error occurred, the device was active or idle.
After command completion occurred, registers were:
ER ST SC SN CL CH DH
04 71 04 9d 00 32 e0 Device Fault; Error: ABRT
Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
a1 00 00 00 00 00 a0 02 00:27:21.427 IDENTIFY PACKET DEVICE
ec 00 00 00 00 00 a0 02 00:27:21.420 IDENTIFY DEVICE
00 00 00 00 00 00 00 04 00:27:21.267 NOP [Abort queued commands]
a1 00 00 00 00 00 a0 02 00:27:20.931 IDENTIFY PACKET DEVICE
ec 00 00 00 00 00 a0 02 00:27:20.887 IDENTIFY DEVICE
Error 97 occurred at disk power-on lifetime: 8426 hours (351 days + 2 hours)
When the command that caused the error occurred, the device was active or idle.
After command completion occurred, registers were:
ER ST SC SN CL CH DH
04 71 04 9d 00 32 e0
Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
ec 00 00 00 00 00 a0 02 00:27:21.420 IDENTIFY DEVICE
00 00 00 00 00 00 00 04 00:27:21.267 NOP [Abort queued commands]
a1 00 00 00 00 00 a0 02 00:27:20.931 IDENTIFY PACKET DEVICE
ec 00 00 00 00 00 a0 02 00:27:20.887 IDENTIFY DEVICE
c8 00 30 af 05 ae e3 00 00:27:17.797 READ DMA
Error 96 occurred at disk power-on lifetime: 8426 hours (351 days + 2 hours)
When the command that caused the error occurred, the device was active or idle.
After command completion occurred, registers were:
ER ST SC SN CL CH DH
04 71 04 9d 00 32 e0 Device Fault; Error: ABRT
Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
a1 00 00 00 00 00 a0 02 00:27:20.931 IDENTIFY PACKET DEVICE
ec 00 00 00 00 00 a0 02 00:27:20.887 IDENTIFY DEVICE
c8 00 30 af 05 ae e3 00 00:27:17.797 READ DMA
c8 00 40 e7 ce 4a e4 00 00:27:17.774 READ DMA
c8 00 10 a7 44 b9 e3 00 00:27:17.747 READ DMA
SMART Self-test log structure revision number 1
Num Test_Description Status Remaining LifeTime(hours) LBA_of_first_error
# 1 Extended offline Completed: read failure 90% 8428 6574443
# 2 Short offline Completed: read failure 90% 8428 184053
# 3 Extended offline Completed: read failure 90% 8428 6574443
# 4 Short offline Completed: read failure 90% 8427 184053
SMART Selective self-test log data structure revision number 1
SPAN MIN_LBA MAX_LBA CURRENT_TEST_STATUS
1 0 0 Not_testing
2 0 0 Not_testing
3 0 0 Not_testing
4 0 0 Not_testing
5 0 0 Not_testing
Selective self-test flags (0x0):
After scanning selected spans, do NOT read-scan remainder of disk.
If Selective self-test is pending on power-up, resume after 0 minute delay.
All errors seem to be same type, some kind of abort signals... Too bad I don't understand anything of this, so I my bag is out of solutions
That "dd_rescue" could work if I had 750G drive, which I dont. I think now my only option is mount the home partition as read only to prevent more damage, and try to copy remaining data in smaller pieces - maybe then the drive won't bork.

V570 - Has too many partitions - Can I nuke one? I need a separate Data Partition!

Hi,
I convinced a Buddy to buy a V570 two days ago. (I've used a Lenovo Laptop for work for a year, and it's been rock solid.) The box was delivered to his house a few hours ago. He lit it up and put his name on it, and then brought it to me, as I help his family out with their computers. (Our Wives are best friends.) I'm a bit geeky and usually know my way around these boxes. But I'm a little stumped on this one.
Normally on laptops they come with tons of bloatware, and it's usually faster to go up to the OEM's site, download all the latest drivers and utilities, then back up the original image before blowing it away, and installing a fresh Windows OS. One of the other main issues is most lappys have one big partition for the OS and Data. So if you get a Trojan or virus => it can nuke all of it. So I always divide the drive, and have a Data partition, so you don't lose everything if (when) something nasty occurs.
But in this case the Box really doesn't have too much bloatware, so I thought make the Factory backup disks, then uninstall the few unwanted programs, partition the C Drive, and create a Data drive in that space. I didn't even look at the hidden partitions so when I went to Create the Data partition I was warned Windows 7 64 Pro doesn't support more than 4 partitions on the operating system drive! If I go into Disk Management I see the typical small Win7 recovery 200 MB partition, a 654 GB C partition, a empty (?) 29 Gig Lenovo partition, and a 14.75 OEM partition. I'm used to the first 2 partitions but what's the rational for the last two from lenovo? One is the "factory" image, and I assume it's the OEM partition, so what's the Lenovo partition for? Can I shrink C, and enlarge the Lenovo partition, so it can become the Data Drive => without affecting the lenovo drives function?
I thought I may have to go with the new Windows install but I can find all the drivers /utilities software at the Lenovo Sites. So I'm a bit stumped and would appreciate your help.
Thanks for the assist.
Regards,
-Jim-

Jim,
     In one word, no.
     I would strongly advise you to just leave it alone, unless you want to spend hours on the phone to your friend, trying to fix all the problems your friend will encounter, with his "new" modded system.
     Please, I'm not trying to be nasty, or unkind. Those 4 partitions are the key to Lenovo's One Key Recovery. Lenovo has spent a lot of time, and engineering to make their systems as user friendly, and trouble free as possible. They built in the Enhanced Experience 2.0, and customized the Windows 7 software for fast booting. Just because you "can" repartition the drive, doesn't mean you "should".
     These forums are filled with users who messed around with their drive structure, used a modded BIOS, or just plain screwed up their programs and hardware, just because they "could". Then they want some urgent help for their machines that went wacky. The moderators on these sites have a hard enough job, keeping track of the various systems "as shipped", to do any troubleshooting on those systems that the user has repositioned the drive, or deleted the programs that Lenovo has included for recovery.
     As you've said, there's not a lot of bloatware on Lenovo's machines. You shouldn't have to reload a "fresh Window OS". If you do, some of the Lenovo's Enhanced Experience 2.0 for Win 7, and some other functions will be toast. That's why you don't see all of the Windows drivers on their site. Lenovo only includes the drivers that have been modified for your V570. Stick with them.
     So, in the end, yeah you can mess with the partitions, but you shouldn't. See this thread to know what you are getting in to, and be ready for your friend's late night phone calls.
http://forum.lenovo.com/t5/IdeaPad-Y-U-V-and-Z-series/new-disk-partioning-and-one-key-recover-featur...
DragonRider
I'm DO'ing IT
Lenovo Y470 085525U 2nd generation Intel® Core™ i7-2630QM processor ( 2.00GHz 1333MHz 6MB )
IdeaCentre A520 Intel® Core™ i5-3230M processor ( 2.60GHz 2600MHz 3MB )
YOGA Tablet 2 Pro-1380F
X1 Carbon 2nd Gen (20A8) ThinkPad + T420

How to fix ORA-01502: index or partition of such index is in usable state?

Greetings,
I received the ORA-01502: index or partition of such index is in un-usable state from one of our Oracle forms, and the problem was caused by using SQL Loader w/ the "DIRECT" load option. As a result of that, the system won't allow any updates including insert/delete to the problematic table w/ the index in un-usable state. I am trying to fix the problem but so far no luck:
When I tried to delete the duplicate rows from the table, I received the ORA-01502: index or partition of such index is in un-usable state error message;
When I tried to rebuild the index by using the following command:
ALTER INDEX index_name REBUILD
I received the "ORA-01452: cannot CREATE UNIQUE INDEX; duplicate keys found
Any help, suggestions or advice on this would be greatly appreciated.
Thanks millions in advance!

The duplicate key rows were inserted to the database by using the DIRECT parameter of the sql loader to specify direct path loading which bypasses integrity constraints. Since the issue exists in our production database and I am primarily an A/P w/ some responsibilities of DBA , I could really use your help to fix our production problem. I would really appreciate if you or any other guru could review my plan and code below for the fixes followed w/ some questions and give me your feedback/comments/suggestions as soon as possible:
1 - Drop the offending index by using the following command:
DROP INDEX <offending index name>
2 - Run the following script to detect duplicate rows:
SELECT *
FROM <Table_Name> A
WHERE
A.rowid >
any(SELECT B.rowid FROM <Table_Name> B
WHERE
B.KEY1 = A.KEY1
and
B.KEY2 = A.KEY2
and
B.KEY3 = A.KEY3
and
B.KEY4 = A.KEY4
3. Run the following script to remove the duplicates:
DELETE
FROM <Table_Name> A
WHERE
A.rowid >
any(SELECT B.rowid FROM <Table_Name> B
WHERE
B.KEY1 = A.KEY1
and
B.KEY2 = A.KEY2
and
B.KEY3 = A.KEY3
and
B.KEY4 = A.KEY4
4 - Rebuild the index by using the following line command:
ALTER INDEX index_name REBUILDBut wait, how about data? Don't I need to back up and reload the data? Do I need to drop and recreate the table? Anything else I should be taking into considerations?
p.s. Although, we only have a total of 35,736 rows of data from this table, but I want to make sure I do it right the first time.
Thanks a lot and any/all the help!

[~SOLVED] 1 min hang during boot while mounting xfs partitions (LVM)

Hi all,
After the upgrade I did yesterday (which included linux-3.12.8-1 and lvm2), the boot time has increased by 1 minute as the system hangs while mounting two partitions formatted as xfs and residing on logical volumes.
$ systemd-analyze blame
1min 13.676s home-VDP76-videos.mount
1min 13.111s home-VDP76-VBox_VMs.mount
2.980s [email protected]
I have the fsck hook as second-to-last entry in my mkinitcpio.conf, but the fstab entries for these partitions include
... xfs defaults 0 0
so, as far as I undertstand, the partitions should not be checked, am I missing something!?
I have googled for similar issues but I did not find anything useful, any suggestion would be greatly appreciated.
Thanks
edit: I blame the check disk because of what reported on the system journal (note the time stamps):
Jan 23 00:10:37 ArchTP kernel: psmouse serio5: alps: Unknown ALPS touchpad: E7=10 00 64, EC=10 00 64
Jan 23 00:10:39 ArchTP kernel: psmouse serio5: trackpoint: IBM TrackPoint firmware: 0x0e, buttons: 3/3
Jan 23 00:10:40 ArchTP kernel: input: TPPS/2 IBM TrackPoint as /devices/platform/i8042/serio4/serio5/input/input25
Jan 23 00:11:47 ArchTP systemd[1]: Mounted /home/VDP76/VBox_VMs.
Jan 23 00:11:47 ArchTP kernel: XFS (dm-4): Ending clean mount
Jan 23 00:11:47 ArchTP kernel: XFS (dm-3): Ending clean mount
Jan 23 00:11:47 ArchTP systemd[1]: Mounted /home/VDP76/videos.
Jan 23 00:11:47 ArchTP systemd[1]: Starting Local File Systems.
Jan 23 00:11:47 ArchTP systemd[1]: Reached target Local File Systems.
I should also add that the involved partitions are ~20 Gb each and quite full...
edit2: I have freed up almost all space on the video partition and the time for mounting those partitions has now decreased to 8 sec...I will mark the thread as solved, but I still would appreciate an input on the reasons why the partitions are checked regardless of the fstab setting, for example, is this pertinent!?
dumpe2fs -h /dev/mapper/waldorf-video | grep -i 'mount count'
dumpe2fs 1.42.8 (20-Jun-2013)
dumpe2fs: Bad magic number in super-block while trying to open /dev/mapper/waldorf-video
Last edited by VDP76 (2014-01-22 23:38:25)

WonderWoofy wrote:I don't think that the fsck.xfs is really a fsck at all. It is there for compatibility purposes, and a real check can be done with xfs_check and xfs_repair. So the fsck.xfs will just always give an exit status of zero.
exactly..in fact the man page says
fsck.xfs(8) wrote:fsck.xfs - do nothing, successfully
so fsck.xfs simply exits with a zero exit status.
running xfs_check gives
xfs_check is deprecated and scheduled for removal in June 2014.
Please use xfs_repair -n <dev> instead.
and xfs_repair does not find errors on both partitions.
Still, I do not understand how nothing being done can take up so much time.
By the way, a short S.M.A.R.T. self-test did not report errors...
In the meanwhile, there has been another kernel upgrade and the timing issue persists, so I am assuming it has something to do with the last lvm2 upgrade.

Mount params to speed up XFS partition

I have Arch installed on a LVM volume using the XFS file system. I'd like to know what mount params I could add to give it a speed boost. This is what I have so far:
# /etc/fstab: static file system information
# <file system> <dir> <type> <options> <dump> <pass>
none /dev/pts devpts defaults 0 0
none /dev/shm tmpfs defaults 0 0
/dev/cdrom /media/cd auto ro,user,noauto,unhide 0 0
/dev/dvd /media/dvd auto ro,user,noauto,unhide 0 0
/dev/fd0 /media/fl auto user,noauto 0 0
/dev/system/root / xfs noatime 0 0
/dev/system/home /home xfs noatime 0 0
/dev/sda5 /boot ext2 noatime 0 0
Now I know that the dump and pass params aren't correct for root. I don't know what to set that to either.
Also how would I hide the physical LVM partition from Windows XP?
Last edited by jordanwb (2008-08-27 22:14:15)

LOL @ the windows-advice!
You know I've been a linux convert since almost two years now (without looking back), and I'm only beginning to scratch the surface of the possibilities with linux, whereas my understanding of the (pre-vista) windows-environment was at a (very) high level. Seriously: this change of mindset (from pro to newbie) was the hardest part of the switch, and only in these last weeks I've been able to contribute to this community with some (easy) answers to problems I had encountered myself.
Glad my windows-knowledge serves someone in here

Mirrorerd boot drive and separate service partition - unsupported ?!?

Hello
i have setup my SL MacMini (2 internal HDs) in a way that I have splitt the harddrives into two parttitions and mirrored them.
Which is possible in the Installer (and even with Lion) . Which gives you the layout shown below. On the seperat partition are also all service data (Mail,Web,...).
From the beginning I had once per month one partition dropping out of the mirror. it was always the same partition/disk having issues.
It was no problem getting it back via: "diskutil appleraid repairmirror 5B12C7D9-89DC-4593-8BEB-28B3F6F6C5FA disk1s2".
As the frequency has gone up i decided to give the MacMini to AppleCare. They found that the motherboard had some issues and exchanged it. But they also mentioned that this way of setting up the drives is unsupported by Apple. You always have to use the full harddrives and mirror them.
Can anyone confirm this ?
For me this sounds stupid. Any process from a user filling up the the homedirectory will make the full system unusable.
Yours
Wieland
P.S.: By exchanging the motherboard the issue didnt go away.
[macmini:~] % diskutil list
/dev/disk0
   #:                       TYPE NAME                    SIZE       IDENTIFIER
   0:      GUID_partition_scheme                        *500.1 GB   disk0
   1:                        EFI                         209.7 MB   disk0s1
   2:                 Apple_RAID                         120.1 GB   disk0s2
   3:                 Apple_Boot Boot OSX                134.2 MB   disk0s3
   4:                 Apple_RAID                         379.5 GB   disk0s4
   5:                 Apple_Boot Boot OSX                134.2 MB   disk0s5
/dev/disk1
   #:                       TYPE NAME                    SIZE       IDENTIFIER
   0:      GUID_partition_scheme                        *500.1 GB   disk1
   1:                        EFI                         209.7 MB   disk1s1
   2:                 Apple_RAID                         120.0 GB   disk1s2
   3:                 Apple_Boot Boot OSX                134.2 MB   disk1s3
   4:                 Apple_RAID                         379.6 GB   disk1s4
   5:                 Apple_Boot Boot OSX                134.2 MB   disk1s5
/dev/disk2
   #:                       TYPE NAME                    SIZE       IDENTIFIER
   0:                  Apple_HFS mirror                 *379.5 GB   disk2
/dev/disk5
   #:                       TYPE NAME                    SIZE       IDENTIFIER
   0:                  Apple_HFS server                 *120.0 GB   disk5
[macmini:~] % diskutil appleraid list
AppleRAID sets (2 found)
===============================================================================
Name:                 server
Unique ID:            8FA04354-2C35-4A13-A7ED-FA74B3F8CE1C
Type:                 Mirror
Status:               Online
Size:                 120.0 GB (119999987712 Bytes)
Rebuild:              automatic
Device Node:          disk5
#   Device Node       UUID                                   Status
0   disk0s2           DA38BBE3-041C-402A-87EC-074FDE9800EC   Online
1   disk1s2           4D77F50D-48E8-4051-BEF4-0A42EED309E8   Online
===============================================================================
===============================================================================
Name:                 mirror_i
Unique ID:            FFEBF2F1-8718-41ED-8032-FD7BE5FA74A0
Type:                 Mirror
Status:               Online
Size:                 379.5 GB (379521794048 Bytes)
Rebuild:              automatic
Device Node:          disk2
#   Device Node       UUID                                   Status
0   disk0s4           57636927-E251-4443-901B-466242D5E4D0   Online
1   disk1s4           DF9D9DC3-C4FE-43D5-836A-5997F9AAD182   Online
===============================================================================

Everything I've seen about about professional apps and looking for extra speed suggests partitioning the boot drive to put your system on the outer, fastest part of the disk and separating your boot/apps and your data.
There was, or is, slightly conflicting info here:
http://macperformanceguide.com/
On the one hand, Lloyd discounted boot drive speed as a factor in day to day work, and emphasized putting DATA on the outer partition, particularly for scratch. However, once he tried a superfast solid state boot drive:
http://macperformanceguide.com/Mac-LightingFast.html
He was sold! Your set-up and purchases clearly indicate performance is your priority. With 4 1 TB drives in RAID, the extra space on the 300 raptor doesn't seem significant for extra data storage.
I looked hard and had trouble finding a definite recommendation for boot partition size or ratio, but 80 GB seems a ballpark number. You want a cushion. If you can guesstimate the potential size of your boot and apps, some of the speed junkies might be able to give you the optimal number. Nice set up, by the way.

HT1449 I want to put different iTunes libraries on separate HD partitions

I want to put iTunes library A on HD partition A and iTunes library B on HD partition B - how do I do this please?

Those files all have different functions. Your "library" is really everything in the iTunes folder.
What are the iTunes library files? - http://support.apple.com/kb/HT1660
More on iTunes library files and what they do - http://en.wikipedia.org/wiki/ITunes#Media_management
What are all those iTunes files? - http://www.macworld.com/article/139974/2009/04/itunes_files.html
Where are my iTunes files located? - http://support.apple.com/kb/ht1391
As for the startup issue you do not change libraries by starting iTunes by clicking on an iTunes file. You must start iTunes from the application icon while holding down the option/alt (not command) key and selecting a library folder or the .itl file..

Bootcamp with Separate Data Partition

I want to create a Bootcamp partition for Win7, and a 3rd partition on the internal HDD for Data. After testing so far, ExFAT is the only format that I've found that both Win7 and OSX can read/write. Is there any reason this setup wouldn't work? Any recommended alternatives? Thanks.

Thanks. Any comments on these posts: https://discussions.apple.com/message/16041384#16041384, https://discussions.apple.com/message/16041377#16041377 and http://www.techparaiso.com/90-steven-paul-jobs-apple-inc/apple-mac-osx/apple-mac -devices/100-best-way-to-setup-osx-mountain-lion-windows-8-data-partition-on-ma c book-air-pro-tutorial? It seems that others have stated that it can be done.
Also, are there tools to make each partition read/write available to the other. Really what I'm trying to get around is having an external drive as a common datastore. I want to keep it all on the internal drive instead.

Separate home partition

Hi, I would like my /home directory on a seperate partition. It is going to be shared with Ubuntu 10.10. I already have it set up in Ubuntu 10.10, but now I need to set it up in Arch Linux. Since all my home contents are already on the new /home partition, I just need to tell Arch Linux to mount that partition automatically.
How can I do just that?
Im assuming I would generally follow this: https://help.ubuntu.com/community/Parti … ome/Moving except for any steps that invlove copying home directories. Is this right?
I just want to make sure, just so I don't screw things up.

TWILisawesome wrote:
Update:
I issued the cat command in both OS's and they report almost the same message. In Ubuntu:
james:x:1000:1000:/home/james:/bin/bash
In Arch:
james:x:1000:100:/home/james:/bin/bash
There is one less zero in the arch command. Could this be a concern?
Any file you create will have its group the same as default group of your user. So, in order to make them similar just change one of these users default group. See man useradd about -g parameter.

How to create a separate /var partition on solaris non-global zone

Hi
I found no simple way to create a separate /var partition in solaris non-global zone.
I am using solaris 10 u9 and my root pool is zfs. My zone's zonepath is also separate zfs fs.
But, I do not know how to make the /var as a mountpoint of another zfs dataset since /var is not empty.
I also do not know if there is a way to install a zone with /var as a separate (outside '/') partition.
That will be really useful.
Any suggestion?
Thanks
Edited by: vadud3 on Sep 20, 2010 12:16 PM

I meant a separate zfs fs with mountpoint '/var' in a non-global zone.
I am insisting, because I do not want /var to fill up the '/' on non-global zone.
With default non-global zone installation, you cannot avoid that.
My zonepath itself is a zfs fs. I also have a zfs dataset provisioned to the non-global zone.
I cannot create a zfs fs out of that dataset and mount it as '/var' becasue by then non-global zone
already installed content on '/var'
I want the '/var' as a separate dir or mountpoint, the same reason global zone gives you that option during install.
Thanks

Systemd error when using separate /var partition

I recently moved /var to a new partition and have noticed this error in journalctl log
Dec 05 07:48:00 archlinux systemd[1]: Started Console System Shutdown Logging.
Dec 05 07:48:00 archlinux systemd[1]: var.mount mount process exited, code=exited status=32
Dec 05 07:48:00 archlinux systemd[1]: Failed unmounting /var.
I added an entry in fstab for /var
/dev/sda4 /var ext4 defaults,noatime,discard 0 1
Drive is ssd and I simply added the options I have for / is this correct?

HungGarTiger wrote:The error message is actually showing a successful unmount by journald not being able to write to the log file....
What? I don't understand.
The log says "umounting /var" then "Failed to umount /var"
What I think falconindy is saying is that, yes systemd will still try and fail to umount /var. However, if you add "shutdown" to the end of your HOOKS="" array in /etc/minitcpio.conf and rebuild the initramfs. That shutdown hook will umount /var after all the systemd and journald stuff is no longer running.
Personaly, I'm just going have journald write the journal to /run/log/journal. I run AIDE before and after every upgrade, so I will know if anything funky happens to my system. A log would be usefull, but as long as I know what files have been changed I can recover them from my backups.
Last edited by hunterthomson (2012-12-11 06:22:36)

No uninitialized partition for /var but I went on anyways

In case anyone makes the same mistake as I did and added the old /var volume back into the new 1.1 VM, this is how you fix it if you told it to keep the /var on the root partition:
(You can do all of this while the VM is running, at least on VMWare)
- Remove old /var disk.
- Create a new disk for the /var volume.
- Log in to the console as root (not ssh)
What you need to know at this point:
When looking at the device nodes for the virtual disks, 0:0 will be /dev/sda, 0:1 will be /dev/sdb and 0:2 will be /dev/sdc.
The order for devices will be 2:1, where the lowest number is always /dev/sda, which tends to be 0:0, so 0:0 (sda) 0:1 (sdb) 1:0 (sdc), the first number is the controller number and the 2nd the disk number.
You can *probably* also check the 'dmsg' output and the new drive will be at the bottom.
Replace 'sdc' with your var drive identifier.
As root:
# fdisk /dev/sdc
Type p just to check if there are no partitions on the drive and you picked the right one.
Type n to create a new partition, make it a primary partition with nr 1, accept the default values and it should work.
exit fdisk by typing w (save and quit)
Now we need to format the new partition:
# mkfs.ext3 /dev/sdc1
(could also make it ext4 but everything else seems to be ext3)
Now comes the trickery to change the var folder around with a mount point.
First we switch to runlevel 1.
# init 1
This will stop most of the services and it will ask for a root pw.
Now we mount the new drive and copy everything over, move the old var out of the way and mount the new drive in its place.
# mkdir /mnt/var
# mount /dev/sdc1 /mnt/var
# cd /var
# cp -ax * /mnt/var
# cd /
# mv var var.old
# mkdir /var
# umount /mnt/var
# mount /dev/sdc1 /var
# ll /var (just to check)
# vi /etc/fstab (or whichever editor you're comfortable with)
Obviously we want to mount the volume again on reboot, so add the following line to your fstab:
/dev/sdc1 /var ext3 rw 0 0
Save and exit and reboot the appliance and you're done.
Obviously it's better to add the new disk *before* starting the vm for the first time but I missed that part in the documentation :-)
Hope this'll help someone else.

If you only deleted the device from the page (an did not wipe/erase it) it will reappear if it connects to the internet before the person restored it or tyrned off the Find feature. See:
http://support.apple.com/kb/TS4006

[fixed] /var on separate xfs partition

Similar Messages

Maybe you are looking for