devel February 2007

devel@lists.centos.org

54 participants
23 discussions

CentOS 5 Beta Initial Test Release
by Lance Davis 05 Mar '07

05 Mar '07

We are about to release an initial internal test release of CentOS 5 beta to our qa team. If folks here want to be a part of that test they need to join the centos-qa list (http://lists.centos.org/mailman/listinfo/centos-qa) and send a message to this list (centos-devel) saying that they would like to be a part of the qa process and join the qa team, with details of relevant experience if you are not known to the core developers. Membership of the centos-qa list will then be approved. Lance -- uklinux.net - The ISP of choice for the discerning Linux user.

34 38

Back-port md raid1/10 BIO_RW_SYNC patch for DRBD
by Ross S. W. Walker 02 Mar '07

02 Mar '07

Attention CentOS Developers, Here is a back-port of Lars Ellenberg's md raid1/10 patch from 2.6.19+ to 2.6.9. It seems the md raid1 and raid10 drivers were stripping the BIO_RW_SYNC flags on requests which caused a severe performance penalty for DRBD when writing it's meta-data to these volumes. The patch is attached as well as inlined: ---------- BEGIN ---------- Subject: md: pass down BIO_RW_SYNC in raid{1,10} From: Lars Ellenberg <[EMAIL PROTECTED]> md raidX make_request functions strip off the BIO_RW_SYNC flag, thus introducing additional latency. Fixing this in raid1 and raid10 seems to be straightforward enough. For our particular usage case in DRBD, passing this flag improved some initialization time from ~5 minutes to ~5 seconds. --- raid1.c | 11 ++++++++--- raid10.c | 11 ++++++++--- 2 files changed, 16 insertions(+), 6 deletions(-) diff -puN drivers/md/raid1.c~md-pass-down-bio_rw_sync-in-raid1 drivers/md/raid1.c --- a/drivers/md/raid1.c~md-pass-down-bio_rw_sync-in-raid110 2007-02-24 01:51:20.000000000 -0500 +++ b/drivers/md/raid1.c 2007-02-24 01:50:38.000000000 -0500 @@ -521,6 +521,7 @@ static int make_request(request_queue_t struct bio *read_bio; int i, disks; mdk_rdev_t *rdev; + const int do_sync = bio_sync(bio); if (unlikely(bio_barrier(bio))) { bio_endio(bio, bio->bi_size, -EOPNOTSUPP); @@ -582,7 +583,7 @@ static int make_request(request_queue_t read_bio->bi_sector = r1_bio->sector + mirror->rdev->data_offset; read_bio->bi_bdev = mirror->rdev->bdev; read_bio->bi_end_io = raid1_end_read_request; - read_bio->bi_rw = READ; + read_bio->bi_rw = READ | do_sync; read_bio->bi_private = r1_bio; generic_make_request(read_bio); @@ -625,7 +626,7 @@ static int make_request(request_queue_t mbio->bi_sector = r1_bio->sector + conf->mirrors[i].rdev->data_offset; mbio->bi_bdev = conf->mirrors[i].rdev->bdev; mbio->bi_end_io = raid1_end_write_request; - mbio->bi_rw = WRITE; + mbio->bi_rw = WRITE | do_sync; mbio->bi_private = r1_bio; atomic_inc(&r1_bio->remaining); @@ -637,6 +638,9 @@ static int make_request(request_queue_t raid_end_bio_io(r1_bio); } + if (do_sync) + md_wakeup_thread(mddev->thread); + return 0; } @@ -960,6 +964,7 @@ static void raid1d(mddev_t *mddev) (unsigned long long)r1_bio->sector); raid_end_bio_io(r1_bio); } else { + const int do_sync = bio_sync(r1_bio->master_bio); r1_bio->bios[r1_bio->read_disk] = NULL; r1_bio->read_disk = disk; bio_put(bio); @@ -974,7 +979,7 @@ static void raid1d(mddev_t *mddev) bio->bi_sector = r1_bio->sector + rdev->data_offset; bio->bi_bdev = rdev->bdev; bio->bi_end_io = raid1_end_read_request; - bio->bi_rw = READ; + bio->bi_rw = READ | do_sync; bio->bi_private = r1_bio; unplug = 1; generic_make_request(bio); --- a/drivers/md/raid10.c~md-pass-down-bio_rw_sync-in-raid110 2007-02-24 01:51:20.000000000 -0500 +++ b/drivers/md/raid10.c 2007-02-24 01:50:38.000000000 -0500 @@ -663,6 +663,7 @@ static int make_request(request_queue_t struct bio *read_bio; int i; int chunk_sects = conf->chunk_mask + 1; + const int do_sync = bio_sync(bio); if (unlikely(bio_barrier(bio))) { bio_endio(bio, bio->bi_size, -EOPNOTSUPP); @@ -747,7 +748,7 @@ static int make_request(request_queue_t mirror->rdev->data_offset; read_bio->bi_bdev = mirror->rdev->bdev; read_bio->bi_end_io = raid10_end_read_request; - read_bio->bi_rw = READ; + read_bio->bi_rw = READ | do_sync; read_bio->bi_private = r10_bio; generic_make_request(read_bio); @@ -789,7 +790,7 @@ static int make_request(request_queue_t conf->mirrors[d].rdev->data_offset; mbio->bi_bdev = conf->mirrors[d].rdev->bdev; mbio->bi_end_io = raid10_end_write_request; - mbio->bi_rw = WRITE; + mbio->bi_rw = WRITE | do_sync; mbio->bi_private = r10_bio; atomic_inc(&r10_bio->remaining); @@ -801,6 +802,9 @@ static int make_request(request_queue_t raid_end_bio_io(r10_bio); } + if (do_sync) + md_wakeup_thread(mddev->thread); + return 0; } @@ -1247,6 +1251,7 @@ static void raid10d(mddev_t *mddev) (unsigned long long)r10_bio->sector); raid_end_bio_io(r10_bio); } else { + const int do_sync = bio_sync(r10_bio->master_bio); rdev = conf->mirrors[mirror].rdev; if (printk_ratelimit()) printk(KERN_ERR "raid10: %s: redirecting sector %llu to" @@ -1258,7 +1263,7 @@ static void raid10d(mddev_t *mddev) bio->bi_sector = r10_bio->devs[r10_bio->read_slot].addr + rdev->data_offset; bio->bi_bdev = rdev->bdev; - bio->bi_rw = READ; + bio->bi_rw = READ | do_sync; bio->bi_private = r10_bio; bio->bi_end_io = raid10_end_read_request; unplug = 1; ----------- END ----------- Ross S. W. Walker Information Systems Manager Medallion Financial, Corp. 437 Madison Avenue 38th Floor New York, NY 10022 Tel: (212) 328-2165 Fax: (212) 328-2125 WWW: http://www.medallion.com <http://www.medallion.com> ______________________________________________________________________ This e-mail, and any attachments thereto, is intended only for use by the addressee(s) named herein and may contain legally privileged and/or confidential information. If you are not the intended recipient of this e-mail, you are hereby notified that any dissemination, distribution or copying of this e-mail, and any attachments thereto, is strictly prohibited. If you have received this e-mail in error, please immediately notify the sender and permanently delete the original and any copy or printout thereof.

3 18

evil centosplus dependency circle
by Michael Gabriel 01 Mar '07

01 Mar '07

trying a yum update on mysql, postgres or php forces the others to install. is that really wanted ??

7 7

centosplus php 5 dependencies
by Michael Gabriel 27 Feb '07

27 Feb '07

is it really wanted to have php 5 depend on mysql 5 AND postgres 8 ? its installing the mysql client 14 for the installed mysql 4.1.20. but why the others ? > yum update php ============================================================================= Package Arch Version Repository Size ============================================================================= Updating: php x86_64 5.1.6-3.el4s1.5 centosplus 1.2 M Installing for dependencies: mysqlclient14 x86_64 4.1.22-1.el4s1.1 centosplus 1.2M php-cli x86_64 5.1.6-3.el4s1.5 centosplus 2.3 M php-common x86_64 5.1.6-3.el4s1.5 centosplus 135 k Updating for dependencies: mysql x86_64 5.0.27-1.el4.centos centosplus 3.3 M mysql-server x86_64 5.0.27-1.el4.centos centosplus 10 M php-devel x86_64 5.1.6-3.el4s1.5 centosplus 489 k php-gd x86_64 5.1.6-3.el4s1.5 centosplus 108 k php-imap x86_64 5.1.6-3.el4s1.5 centosplus 50 k php-ldap x86_64 5.1.6-3.el4s1.5 centosplus 33 k php-mbstring x86_64 5.1.6-3.el4s1.5 centosplus 990 k php-mysql x86_64 5.1.6-3.el4s1.5 centosplus 85 k php-ncurses x86_64 5.1.6-3.el4s1.5 centosplus 39 k php-odbc x86_64 5.1.6-3.el4s1.5 centosplus 50 k php-pdo x86_64 5.1.6-3.el4s1.5 centosplus 228 k php-pgsql x86_64 5.1.6-3.el4s1.5 centosplus 66 k php-snmp x86_64 5.1.6-3.el4s1.5 centosplus 26 k php-xmlrpc x86_64 5.1.6-3.el4s1.5 centosplus 52 k postgresql x86_64 8.1.8-1.el4s1.1 centosplus 2.8 M postgresql-libs x86_64 8.1.8-1.el4s1.1 centosplus 192 k postgresql-server x86_64 8.1.8-1.el4s1.1 centosplus 4.0 M

2 1

Re: [CentOS-devel] Joining QA mailinglist
by Patrick 27 Feb '07

27 Feb '07

>> Last week, I posted my request here to join the QA team. >> Also, I signed up with the QA mailinglist. >> >> I received a confirmation after which I needed to review my settings and submit them for further processing. After that, I did not hear anything. >> >> Anyone else got the same problem? >> Should I resubmit? > >We are just being slow at approving people ... > > >Regards >Lance > Thanks for the info. I almost started to worry. Regards, Patrick _______________________________________________ No banners. No pop-ups. No kidding. Make My Way your home on the Web - http://www.myway.com

1 0

Joining QA mailinglist
by Patrick 27 Feb '07

27 Feb '07

Last week, I posted my request here to join the QA team. Also, I signed up with the QA mailinglist. I received a confirmation after which I needed to review my settings and submit them for further processing. After that, I did not hear anything. Anyone else got the same problem? Should I resubmit? Regards, Patrick _______________________________________________ No banners. No pop-ups. No kidding. Make My Way your home on the Web - http://www.myway.com

2 1

CentOS 4 Samba - Excel 2002/2003 bug
by Rodrigo Barbosa 26 Feb '07

26 Feb '07

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 There is a nasty bug on samba, up to 3.0.10, which causes Excel to incorrectly think a file was changes by someone else. This is a know issue, fixed on samba 3.0.11 (stock). A patch was also applied to samba on RHEL 3, but was NOT on RHEL 4. This patch (samba-3.0.9-excel.patch) will apply cleanly on samba-3.0.10-1.4E.9, and build without a glinch. This issue is NOT solves by http://support.microsoft.com/kb/324491 (has it on a machine with Office SP3 installed). Maybe we can get a centosplus version of samba with this patch applied ? []s - -- Rodrigo Barbosa "Quid quid Latine dictum sit, altum viditur" "Be excellent to each other ..." - Bill & Ted (Wyld Stallyns) -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) iD8DBQFFz7+rpdyWzQ5b5ckRAnX9AKCVm1x5fxyTdrjFa8kLzHLCft7tvACffTJa kpT1XyvlDjOdMoI12djA/h0= =6b5J -----END PGP SIGNATURE-----

5 6

missing mailx bugfix update in upstream mirrors
by Matthew Miller 26 Feb '07

26 Feb '07

See: <https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=184139> The src.rpm for this update for RHEL4 was never pushed properly, but is now where it should be. <http://rhn.redhat.com/errata/RHBA-2006-0687.html> And for some bizarre reason, it's marked as "Affected Products: Other", but it is indeed a RHEL4 fix (and cures an issue that bites us here). Probably you guys will notice this yourselves, but I figured I'd mention it just in case. Thanks. -- Matthew Miller mattdm(a)mattdm.org <http://mattdm.org/> Boston University Linux ------> <http://linux.bu.edu/>

2 2

Kernel used on installation media.
by Ray Van Dolson 24 Feb '07

24 Feb '07

Hi all, I hope this is not too off topic. I've searched the -devel archives, but haven't found clear-cut information on this, so thought I'd ask. I am trying to determine the kernel configuration used to build the kernel used on the installation media (CD #1 or the vmlinuz/initrd.img files in their respective images/boot.iso or images/diskboot.img files). My guess is the kernel configurations are not identical to the ones ultimately used to build the kernels used on the system once it is up and running.... I ask because I am interested in rebuilding the kernel used on CD #1 to boot into the installer. Thanks for any pointers. Ray

7 11

Re: [CentOS-devel] Installation RHEL5 - > CentOS5
by Patrick 24 Feb '07

24 Feb '07

Well, since I am new to this mailinglist, I did not know it was a discussion already held (hadn't sniffed the archieves). But thanks for the update! Regards, Patrick --- On Sat 02/24, Karanbir Singh < mail-lists(a)karan.org > wrote: From: Karanbir Singh [mailto: mail-lists(a)karan.org] To: centos-devel(a)centos.org Date: Sat, 24 Feb 2007 07:33:02 +0000 Subject: Re: [CentOS-devel] Installation RHEL5 - > CentOS5 Patrick wrote: > When I installed RHEL5 beta 2 yesterday, I came across the so-called installation number. It's used as a "serial" to unlock number-related functionality of the software. > > In other words, there is a key for the client version and one for the server version. And within those options, there are also different possibilities, like with or without visualisation. > > How does this affect CentOS5? it wont > Will this be implemented or modified? it wont. either implemented or mordified. This is again a conversation best had once the centos beta trees are public. -- Karanbir Singh : http://www.karan.org/ : 2522219@icq _______________________________________________ No banners. No pop-ups. No kidding. Make My Way your home on the Web - http://www.myway.com

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

devel February 2007