Skip to contentSkip to conversation

Recomended IDE-RAID MB

DISCUSSION Andrew B. Cramer 5 participants 8 messages Apr 6, 2003 – Apr 7, 2003

lore
Share & tools
lore
#0 lore raw

Hi All,
What brand of motherboard would you use for just mirroring two huge
IDE drives?

Best - Andrew

#1 lore raw
Quoted 3 lines · Andrew B. Cramer

What brand of motherboard would you use for just mirroring two huge
IDE drives?

I'd just make sure the drives were fast, that the MB supported ATA100 and that you used 80w IDE cables. My Soyo Dragon Plus gets fairly good throughput; unfortunately, my drives are too slow, which means the total performance would be better than:

Version 1.02c ------Sequential Output------ --Sequential Input- --Random-

                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
blackjesus     512M  7128  64  8339   6  4925   3  8256  70 13530   5 161.1   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP /sec %CP
                 16 11707  97 +++++ +++ 13882  99  8859  76 +++++ +++ 12141 100

In my experience it takes at the most 5 minutes to resync the (on-average) 2GB partitions it is cut up into.

This board also includes Promise IDE Hardware RAID, but I've never used it and can't comment on its performance.

Take care,
--
Christian Reis, Senior Engineer, Async Open Source, Brazil.
http://async.com.br/~kiko/ | [+55 16] 261 2331 | NMFL

#2 lore raw

Christian,
Thank you for your input. Just about any speed drive will eat up a T-
1. I'm just looking for maybe most supported, and reliable. Drives are cheap and I will be getting either ATA100 or 133 drives.

Best - Andrew

Quoted 34 lines · Christian Reis
Quoted 3 lines · Andrew B. Cramer

What brand of motherboard would you use for just mirroring two huge
IDE drives?

I'd just make sure the drives were fast, that the MB supported ATA100 and that you used 80w IDE cables. My Soyo Dragon Plus gets fairly good throughput; unfortunately, my drives are too slow, which means the total performance would be better than:

Version 1.02c ------Sequential Output------ --Sequential Input- --Random-

                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
blackjesus     512M  7128  64  8339   6  4925   3  8256  70 13530   5 161.1   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP /sec %CP
                 16 11707  97 +++++ +++ 13882  99  8859  76 +++++ +++ 12141 100

In my experience it takes at the most 5 minutes to resync the (on-average) 2GB partitions it is cut up into.

This board also includes Promise IDE Hardware RAID, but I've never used it and can't comment on its performance.

Take care,
--
Christian Reis, Senior Engineer, Async Open Source, Brazil.
http://async.com.br/~kiko/ | [+55 16] 261 2331 | NMFL
-
To unsubscribe from this list: send the line "unsubscribe linux-admin" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#3 lore raw

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

In my experience where I work, the Tyan motherboards with built in RAID work well.
Some of them have Promise RAID chip, and the higher end ones have Adaptec RAID chip, integrated on the motherboard.

I have no throughput test numbers to give.

Quoted 51 lines · Andrew B. Cramer

Christian,
Thank you for your input. Just about any speed drive will eat up a T-
1. I'm just looking for maybe most supported, and reliable. Drives are cheap and I will be getting either ATA100 or 133 drives.

Best - Andrew

Quoted 27 lines · Christian Reis
Quoted 4 lines · Andrew B. Cramer

What brand of motherboard would you use for just mirroring two huge
IDE drives?

I'd just make sure the drives were fast, that the MB supported ATA100 and that you used 80w IDE cables. My Soyo Dragon Plus gets fairly good throughput; unfortunately, my drives are too slow, which means the total performance would be better than:

Version 1.02c ------Sequential Output------ --Sequential Input- --Random-

                   -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
blackjesus     512M  7128  64  8339   6  4925   3  8256  70 13530   5 161.1   0
                   ------Sequential Create------ --------Random Create--------
                   -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
             files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP /sec %CP
                16 11707  97 +++++ +++ 13882  99  8859  76 +++++ +++ 12141 100

In my experience it takes at the most 5 minutes to resync the (on-average) 2GB partitions it is cut up into.

This board also includes Promise IDE Hardware RAID, but I've never used it and can't comment on its performance.

Quoted 11 lines

Take care,
--
Christian Reis, Senior Engineer, Async Open Source, Brazil.
http://async.com.br/~kiko/ | [+55 16] 261 2331 | NMFL
-
To unsubscribe from this list: send the line "unsubscribe linux-admin" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Quoted 7 lines

-
To unsubscribe from this list: send the line "unsubscribe linux-admin" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

- --

- -----------
73 de K6LNX

Matt M.
LinuxKnight

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (MingW32)

iD8DBQE+kKZS84ta5vs8jhYRArGGAKDUsbfohx+31aN8ZwKeaCsd4wOk9QCgxuiT /fIvvaBwWRdl3r9Ofq2WvMo=
=gh+2
-----END PGP SIGNATURE-----

#4 lore raw

Thanks,
I like TYAN except they have been quite restrictive with memory
requirements before. I'll look at their products again. I think the AHA-RAID chip makes more sense.

Cheers - Andrew (ICQ # 243211228)

Quoted 95 lines · Matt McKenzie

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

In my experience where I work, the Tyan motherboards with built in RAID work well.
Some of them have Promise RAID chip, and the higher end ones have Adaptec RAID chip, integrated on the motherboard.

I have no throughput test numbers to give.

Quoted 52 lines · Andrew B. Cramer

Christian,
Thank you for your input. Just about any speed drive will eat up a T-
1. I'm just looking for maybe most supported, and reliable. Drives are cheap and I will be getting either ATA100 or 133 drives.

Best - Andrew

Quoted 27 lines · Christian Reis
Quoted 4 lines · Andrew B. Cramer

What brand of motherboard would you use for just mirroring two huge
IDE drives?

I'd just make sure the drives were fast, that the MB supported ATA100 and that you used 80w IDE cables. My Soyo Dragon Plus gets fairly good throughput; unfortunately, my drives are too slow, which means the total performance would be better than:

Version 1.02c ------Sequential Output------ --Sequential Input- --Random-

                   -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
blackjesus     512M  7128  64  8339   6  4925   3  8256  70 13530   5 161.1   0
                   ------Sequential Create------ --------Random Create--------
                   -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
             files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP /sec %CP
                16 11707  97 +++++ +++ 13882  99  8859  76 +++++ +++ 12141 100

In my experience it takes at the most 5 minutes to resync the (on-average) 2GB partitions it is cut up into.

This board also includes Promise IDE Hardware RAID, but I've never used it and can't comment on its performance.

Quoted 11 lines

Take care,
--
Christian Reis, Senior Engineer, Async Open Source, Brazil.
http://async.com.br/~kiko/ | [+55 16] 261 2331 | NMFL
-
To unsubscribe from this list: send the line "unsubscribe linux-admin" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Quoted 7 lines

-
To unsubscribe from this list: send the line "unsubscribe linux-admin" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

- --

- -----------
73 de K6LNX

Matt M.
LinuxKnight

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (MingW32)

iD8DBQE+kKZS84ta5vs8jhYRArGGAKDUsbfohx+31aN8ZwKeaCsd4wOk9QCgxuiT /fIvvaBwWRdl3r9Ofq2WvMo=
=gh+2
-----END PGP SIGNATURE-----

-
To unsubscribe from this list: send the line "unsubscribe linux-admin" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#5 lore raw

Kernel Panic

In reply to Andrew B. Cramer

Dear Linux Admin,

On Linux 2.4 I have had three Kernel Panics recently.

They include the following: journal.o "( ret != 0 )". I see in the source code that there is an assertion: J_ASSERT(ret != 0) in the function:

unsigned long journal_bmap(journal_t *journal, unsigned long blocknr)

Which looks like a possible place for the origin of this.

There is also a large amount of Hex data, which I do not have, unless logged somewhere on my system unknown to me. (I don't get many Panics, not sure where panic data is logged...)

On reboot I loose files. Although nothing serious yet.

The system is a POP3 server, handling about 1,000 email files and about 10,000 email's a day, so there is constantly great disk load by sendmail and the default redhat POP3.

Should I upgrade my kernel, or may I be doing something specific to cause this?

Thanks with any help,

Ben Clewett.

#6 lore raw

Re: Kernel Panic

Sounds to me like you have bad sectors on your harddisk that the journalling file system is trying to work around. What file system are you using and what hardware?

cheers

Jamie...

Quoted 33 lines

Dear Linux Admin,

On Linux 2.4 I have had three Kernel Panics recently.

They include the following: journal.o "( ret != 0 )". I see in the source code that there is an assertion: J_ASSERT(ret != 0) in the function:

unsigned long journal_bmap(journal_t *journal, unsigned long blocknr)

Which looks like a possible place for the origin of this.

There is also a large amount of Hex data, which I do not have, unless logged somewhere on my system unknown to me. (I don't get many Panics, not sure where panic data is logged...)

On reboot I loose files. Although nothing serious yet.

The system is a POP3 server, handling about 1,000 email files and about 10,000 email's a day, so there is constantly great disk load by sendmail
and the default redhat POP3.

Should I upgrade my kernel, or may I be doing something specific to cause this?

Thanks with any help,

Ben Clewett.

-
To unsubscribe from this list: send the line "unsubscribe linux-admin" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 1 line
**  This message was transmitted on 100% recycled electrons **
#7 lore raw

Re: Kernel Panic

The filesystem are ext2 and ex3:

Filesystem    Type   1M-blocks      Used Available Use% Mounted on
/dev/hda3     ext3       37038      1860     33296   6% /
/dev/hdc1     ext3        1969       119      1750   7% /apache
/dev/hda1     ext2          30         6        23  19% /boot
/dev/hdc3     ext2       37560       279     35373   1% /var/spool/mail
/dev/hdc4     ext2       35087        76     33229   1% /home

The kernel panic mensions journal.o so I guess this must be an ext3 partision? Probably on the '/' as no files in the 'apache' are ever effected.

I am not sure about the hardware, it's not easy to check, but I belive my /dev/hda is a 40Gb someting purchaced a year back. On IDE. Either a Seagate, a Fujisu or a Western Digital. (Can't get cover off as rack-mounted and rack screwed into removable cover :)

I hope you can help me, as the implications of an old kernel or an old disk have radically different fix's :)

Ben

Quoted 45 lines · Jamie Harris

Sounds to me like you have bad sectors on your harddisk that the journalling file system is trying to work around. What file system are you using and what hardware?

cheers

Jamie...

Quoted 33 lines

Dear Linux Admin,

On Linux 2.4 I have had three Kernel Panics recently.

They include the following: journal.o "( ret != 0 )". I see in the source code that there is an assertion: J_ASSERT(ret != 0) in the function:

unsigned long journal_bmap(journal_t *journal, unsigned long blocknr)

Which looks like a possible place for the origin of this.

There is also a large amount of Hex data, which I do not have, unless logged somewhere on my system unknown to me. (I don't get many Panics, not sure where panic data is logged...)

On reboot I loose files. Although nothing serious yet.

The system is a POP3 server, handling about 1,000 email files and about 10,000 email's a day, so there is constantly great disk load by sendmail and the default redhat POP3.

Should I upgrade my kernel, or may I be doing something specific to cause this?

Thanks with any help,

Ben Clewett.

-
To unsubscribe from this list: send the line "unsubscribe linux-admin" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html