Is that the fast as hell or slow as hell kind?
Slow.
I ask because I have a couple of IBM M5015 (rebranded LSI 9260-8i) controllers
that
I run in RAID-10 and as I'm somewhat on a budget for disks (coming home with
$1200
worth of 15k SAS drives will get me shot) I'd rather run a RAID-5/6 array
I think SATA hard drives are so cheap nowadays, that it's hard to justify
choosing
RAID 5 for more yield at the cost of performance and reliability.
Another point to consider with slower sata discs that are also large is the
prolonged
rebuild time and degraded redundancy during this rebuild,
I have a DRBL server, basically an nfs fileserver, which I am rebuilding. I
want to
put the high i/o directories on a separate raid array for performance.
Currently
everything is under / in one raid array.
How can I tell which directories, obviously other than /home, are getting high
reads
I have several CentOS 6 boxes that mount iscsi based luns and use mpath.
They all had problems shutting down as a result of unused maps not getting
flushed as the system halted.
After examining the init scripts, netfs, iscsi and multipathd all had the
correct
order but mpath failed to flush
3) flush multipaths
4) stop iscsi
I guess that's the point, it seems the init script does not flush them out so
the module and any dependent dm mods stay active.
jlc
___
CentOS mailing list
CentOS@centos.org
Is is possible to use kickstart file to install rhel from dvd drive?
Mainly idea is to clone one anaconda.ks file to about twenty
machines.?
examples? ftp/http/dhcp is not possible due to network limitations.
Of course. Check the manual, it specifies places to look when you specify
the ks file in
Boot from the DVD/ISO, press tab at the first install screen, and
point to your ks file with ks=,,,.
Right, but that doesn't answer his question. The op does not have
ip connectivity:
Afaik the LEDs won't work out-of-the-box.
Maybe,
I presume its vendor specific but if they appear to linux in sys,
maybe a solution like the first link suggests?
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=560432
http://lkml.org/lkml/2008/2/3/193
Can't find what I'm doing wrong
Probably we won't be able to either unless you elaborate on how
you begin the installation:)
Given you posted dhcpd config, I am willing to guess you are
doing|assuming something wrong...
how can I go about extracting the IP address from with in the %pre section
I was going to do something like:
cat /proc/cmdline | tr ' ' '\n' | grep ks=http | tr '/' ' ' | awk {'
print $2'}
Since tr isnt available how else can I do it?
Jerry,
You ever solve this? You can do this all with awk
The thing is if it's in the %pre section, AWK would not have
been installed yet?
I wouldn't have suggested it if it would not have worked.
Awk *is* available in %pre. As the op noted, tr is not available
but several tools like awk,gawk,grep,fgrep,egrep,wget etc are.
Since tr isnt available how else can I do it?
Check out page 21 of this pdf:
http://www.redhat.com/promo/summit/2010/presentations/summit/decoding-the-code/wed/cshabazi-530-more/MORE-Kickstart-Tips-and-Tricks.pdf
Interesting, while the awk solution is more appropriate for your
need, this can
As smartd isn't running in the default Centos configuration, where does
palimpsest get its information? Is it a self-contained program that doesn't
require smartd or is something else happening behind the scenes?
I think you misunderstood the first reply: smartd, as in the init script is a
Which still doesn't answer my question. Perhaps I'm wording it poorly -- I'll
try again:
Perhaps palimpsest runs smartctl and queries the device itself? Perhaps it
borrowed
code from the project and runs the query itself? I don't have any servers with
GUI's,
couldn't tell you...
I did not move anything; it is a brand new install...
The only thing that might be out of ordinary is that the yum update
is made from a kickstart chrooted post script...
You don't need to do that. Add the updates repo as a 'repo' line
in the ks, it will install anything new at once saving time
I'm looking for CentOS 5.0 x86_64 dvd iso (preferably not torrent); must be 5.0
All my web searches have come up with more recent versions of 5.x, not 5.0.
Can someone give me a URL I can wget with? I'd even be willing to
throttle down the wget).
How about you keep the wget unthrottled:)
What was worse was that booting back to 5.6 left most of the raid pairs
broken, some still having numbers in the md120 range, and one of the
drives with an unrecognizable partition table. I think I can repair
everything, but did I miss something about this in release notes somewhere?
I just
for VERSION in 5.6 6.0
Check your httpd error log for a 404:)
Just a WAG, I presume you modified your 5x installs to not look in the default
5 location, where point releases get symlinked too once they are current.
Your 6.0 install looks for a 6 directory.
It does increase server
load and rsync run time, though, so you should probably disable it
again afterwards.
That's why most repo's don't allow the use of it, I use the same master
as you Jerry, and they are certainly far too large to allow that.
___
Anyone have any suggestions?
You thought about using its built in load balancing to scale out the capacity?
Don't know what hardware you are using but it might be worth looking at the
lease db and log file setups you are using if you can't add iron to the
equation.
Anyone else notice that /usr/libexec/bacula/make_catalog_backup.pl runs a
dbcheck -B
which doesn't seem to exist in the version of dbcheck shipped with it?
# strings /usr/sbin/dbcheck |grep -i print
sprintf
vfprintf
snprintf
-dt print timestamp in debug output
-?
Any time line of availability of Asterisk binaries on CentOS version 6.
It might make sense to ask the guys who actually build them?
http://packages.asterisk.org/rhel
http://packages.atrpms.net/dist/el6/asterisk/
___
CentOS mailing list
A) Can CentOS 6 handle that many JBOD disks in one system? is my upper
size too big and I should plan for 2 or more servers? What happens with
the device names when you've gone past /dev/sdz ?
Dev names double, sdaa etc.
B) What is the status of large file system support in CentOS 6? I know
Should I assume that these .iso's are unreliable, and d/l
newer ones?
No, normal behavior.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
Ok, now you have me confused: should there not be such image files?
Look at the file names, it should be intuitive. As per the docs, if they exist,
updates.img will apply updates:) etc...
http://www.arrfab.net/blog/?p=305
Looks like it's been removed from the 6x docs, but I haven't the time to
I've got two gigabit ethernet interfaces bonded in CentOS 5.6. I've
set miimode=1000 and I've tried mode= 0, 4 and 6. I've not been able
to get better than 112MB/sec, which is the same as the non-bonded
interfaces.
Most people misunderstand bonding/link aggregation. A single conversation
I plan on creating CentOS 6.0 Desktop off-spin, changing only release
package to add priorities and enable Plus and Extras repositories
Let me know how that Extra repo addition goes:)
___
CentOS mailing list
CentOS@centos.org
So according to the Bind arm, `rndc freeze zone ...` should drop the jrn file
and
flush its deltas. I have tested this and it's not updating the master on my 5.6
box
with bind-9.3.6, it drops the jrn and discards the deltas after an
unfreeze|thaw.
A service restart behaves as expected.
Anyone
I am trying to install CentOS 5.6 on HP DL 180 G6 which has HP Smart
Array B110i SATA RAID Controller Driver. I have 4 * 500 GB SATA HDD
and configured RAID 1+0 using System BIOS, the BIOS detects usable
disk space as 940GB disk space, when i start installing the OS, it
does not detect HP Smart
Please help me understand.
If the device requires an additional driver, unless its packaged as a dd for
use at
install, how can you install and then add a driver?
Disable RAID mode, set it to AHCI, then Anaconda will see all the individual
discs
at which point during install you can choose to
http://www.ibm.com/developerworks/library/l-shobj/
Thanks, that was informative.
As Ljubomir said, libasync_wmi_lib.so.0 will be a symlink to
libasync_wmi_lib.so.0.0.1. You can use objdump to find the soname(s) for
shared objects, e.g:
objdump -p /path/to/libasync_wmi_lib.so.0.0.1 | grep
I whipped up a quick spec to package Zenoss' wmi client and noticed it creates a
libasync_wmi_lib.so.0 and libasync_wmi_lib.so.0.0.1 that their python script
would
otherwise copy during its install routine (obviously I wrote a native rpm
section for
installation).
Anyone know the naming
Dunno but Google 'sed only first match' gives some answers...
I resorted to just using a perl one liner. It wasn't as straight forward as
that. Using
sed's {p;q;} syntax with the pattern provides the functionality, but it was the
capture
group only I was interested, and not the whole line
Hey guys,
I am trying to extract a pattern from a downloaded html file. I only want the
first match
printed, problem is I don't know how to force sed to exit after printing the
first match of
a capture group. Anyone know the trick to manipulating {p;q;} to print a
capture group?
Thanks,
jlc
I would like to reset the table so that
disk is the USB drive and delete the data that is now on my
main drive.
It's been a while since I paid attention to this, but afaik the directory
under media is the volume name of the disc, with the exception you
found, if that directory exists. Move it
Hey guys,
Need an opinion on what to use for file versioning text conf files that get
updated by
scheduled rsync's etc. Need something that can watch the file, so it doesn't
need an
explicit checkin, and can be diffed by a web front end.
I haven't any preference on backend either.
Thanks for
Git and Gitweb?
Thought of that, is there anything that can monitor for changes so I can
avoid a commit command for every script, as they all dump to an already
well organized tree, I was hoping to monitor the top level dir for changes
and have it commit as they appear.
Something like that
in which file is the %prep script defined?
Defined in the spec file.
www.rpm.org says its a build-in script. Does that mean its defined in
some binary? I wonder how %prep handles the setup of locale.
I would recommend you pull down a few srpms and unpack them
then look at their spec files,
Actually I'm trying to figure out where these lines originate from
I see
Executing(%prep): /bin/sh -e /var/tmp/rpm-tmp.86989
+ umask 022
+ cd /home/user/rpmbuild/BUILD
+ LANG=C
+ export LANG
The setting of LANG messes up my build, which relies on Utf8.
So export this in the build environments
I have Sangoma A101DE 1 Port E1/T1 With Hardware Echo Cancellation ( PCI
Express )
Card installed on the box. Its not detected. Details are as below :-
I have seen these type of cards detected as something else. Its also been ages
since I used
asterisk anyway, I use FreeSWITCH now and as you
Well, Nagios is more for outage/alerts were I was looking more for
trending.
Thats why I was originally using Cacti w/Nagios plugin.
There are plenty of graphing/trending addons for Nagios, I moved to
Icinga and use PNP4Nagios with it.
___
CentOS
Even so, the card with no driver would be listed in lspci...
Sort of, it may be incorrectly identified as something else, all
those pretty names are simple mappings in a text file. See
/usr/share/hwdata/pci.ids
But then again, multiple people have suggested the same suggestions
and/or hints over
For a long period of time, my Apache root directory has been
/home/httpd. For security reasons, this is not so good as
SELinux has informed me. Now all of the files have been copied
to /var/www/etc with owner and group root. The privileges are
754 (rwxr-xr--), however apache does not have
I just did a test install off a netinstall iso to a mirror repo which left
no /etc/sysconfig/iptables file in place.
So a quick check:
# yum whatprovides /etc/sysconfig/iptables --disablerepo=\* --enablerepo=base
--enablerepo=updates --disableplugin=\*
No Matches found
Without that file
And since that file contains the rules list for iptables, it doesn't
make sense for iptables to run without it.
Basically, if you want to run iptables, you need to define some rules
first. Either create the file by hand and then start iptables, or
create the rules in iptables and use service
You can boot ESXi from a
small CF card, as once its booted, it doesn't touch the boot device at all.
Yes it does, there are cron jobs for config backups etc.
How does it remember config changes in a non-stateless deployment?
~ # cat /var/spool/cron/crontabs/root
#syntax : minute hour day
Use fence_manual ... but it is not supported officially by TUV in
production environments ...
Do not use that...
How does that actually fence the offending node when it needs to?
The concept of fencing is for the preservation of data integrity, circumventing
it with a manual fence is useful
Tried sending this a while ago, but looks like mailman blocked it?
But i don't know how to write the spec file so a simple yum update will
install both packages and it won't create any problems.
A few options here, create a %{name} .spec with:
# This is part one, or the main component
I am trying to deduce the one liner for an rpm I am packaging to edit a php
configuration file from within my spec. The line I am editing looks like:
$conf['nagios_base'] = /nagios/cgi-bin;
What are the escape requirements Perl needs from within the shell to
search for this?
As I have seen, the
$conf['nagios_base'] =
I'd just search for that part, above.
Me to, and I never even got to the replacement as the search for that
was bailing:)
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
The problem is trying to pass valid Perl though the bash shell . There
is an insane amount of interaction between all the escapings involved in
this specific pattern. The hard problem is getting bash to *not* change
what you are passing to Perl before Perl sees it.
As I have noticed:)
After
Go easy on yourself, and :
a) don't try to match the quotation marks verbatim. Instead, match them with
simply a .
Clever, thanks!
jlc
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
In fact, you can do things very easily with *nix acls that are very
difficult in Windows. For example, you can set different 'Default'
permissions (what will be on things created in the directory) than the
permissions that are actually on the directory. You can set different
masks for different
*OR* as a special case, if access is *only* read-only (or read-only to
all but one initiator).
Special is large understatement, in almost every case the view of the
ro fs won't match the rw system as soon as it begins writing...
___
CentOS mailing list
I'm trying to create a script, that watches the changes in
/dev/shm/dhcpacks-in-last-2min.txt file [in every 1 sec].
Have look at Gamin, its designed for this type of thing...
___
CentOS mailing list
CentOS@centos.org
Has anyone tried using an LSI MegaRAID 320-2E PCIe RAID Controller in
an HP DL380G5 (of any HP G5 DL for that matter) running CentOS 5? It
looks like it should work but it would be nice to know that someone
has already done it before I start going to all the effort and expense
of building the box.
Any idea what it might be for?
Procmail...
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
I have a WD Elements 2TB USB hard drive. I would like to
encrypt the whole drive with TrueCrypt on my CentOS box.
Could someone direct me to a tutorial on how to do this?
Ideally, I would also like to access it with my WinXP box,
but this part is optional.
Last time I checked I was
The current version of anaconda supports that, but to the best of my
recollection, the version used in RHEL 5 did/does not.
Your recollection is wrong, I have never done a CentOS install except
for the first couple when I was learning without it...
___
[root@gw5fl 2.0]# . ./vars
bash: /usr/share/doc/openvpn-2.1.4/easy-rsa/2.0/whichopensslcnf:
Permission denied
See that error above? Make that script executable... It's a bash script
`vars` is calling and not able to execute.
___
CentOS mailing list
Thanks, did that though this did not fix my problem - I still get the
same error message.
The only error I saw was a lack of ability to run whichopensslcnf.
pkitool is a shell script which should be executable and in that folder as
well. Is it executable?
I guess what I was asking for is to take a already configured server and put
it on multiple CD's DVD's and then use that to install on another server.
Given that that will take some leg work, I accomplish the same thing by making
a long %post section in a kickstart that runs various sed commands
You can have kickststart run a yum update so each box will have the
latest updates as at install time.
Over and over again I see this reco and it makes no sense? If you have
access to updates whether they be yours locally cached or remote, you
should add a repo line in your ks and install updates
I have a WD Elements 2TB USB hard drive. I would like to
encrypt the whole drive with TrueCrypt on my CentOS box.
Could someone direct me to a tutorial on how to do this?
Ideally, I would also like to access it with my WinXP box,
but this part is optional.
Truecrypt is platform agnostic, see the
I am looking at LDAP module for Apache httpd for authentication. The
'yum install' gives me 'mod_authz_ldap.i386 0:0.26-9.el5_5.1', whereas
on Apache documentation site I find mod_authNz_ldap module. Both
^
I am looking at LDAP module for Apache httpd for authentication. The
'yum install' gives me 'mod_authz_ldap.i386 0:0.26-9.el5_5.1', whereas
on Apache documentation site I find mod_authNz_ldap module. Both
modules appear to be different looking at available directives. Any
clues or suggestions on
Can anyone please steer me in the right direction with this one? I've
searched the net, but couldn't find a clear answer.
I have done it before and googled iozone graphing and got quite a few
references?
http://www.linux.com/archive/feature/139744
On Centos 5.5, is there a way to skip the countdown
screen 'booting xyz OS in x seconds', and make GRUB show the
menu of all the bootable OSes, and just stop there?
First hit on Google:)
Comment out the timeout line...
jlc
___
CentOS mailing list
I used the orangeHRM tool , but found it did not have the LDAP authentication .
You sure about that?
http://www.orangehrm.com/wiki/index.php/33%29_How_the_ldap_works_in_OrangeHRM.
___
CentOS mailing list
CentOS@centos.org
Can someone guide me on how to load the drivers so the OS can see the
controller as it loads.
Download the driver, you get a dd disk for new installs when your root boots
off the controller
and an rpm can you use after, and or if the controller is not backing your root.
I am sorry but I didn't understood much of your comment, can you brake it down
further step by step...I am doing baby steps in centos.
Are you installing CentOS to boot off this controller, yes or no?
___
CentOS mailing list
CentOS@centos.org
I wish this can be done from a USB stick since the machine does not have a
floppy.
The dd can be loaded via http, even easier than usb...
Use a kickstart, `driverdisk --source=http://path/to/dd.img`
Something like that is prolly avail from the grub boot menu, see the manual.
Is there any known issues when trying to run CentOS (x86 or x64) on a fresh
install of vmware 4.1?
Details as to why I am asking are here in the CentOS
forums: https://www.centos.org/modules/newbb/viewtopic.php?viewmode=flatorder=DESCtopic_id=28521forum=39
Vmware, no. I have been using it for
Is there an equivalent in CentOS?
cat /etc/passwd |cut -d : -f 3 |sort -n
;)
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
I've always wondered what the 1: means at the beginning of a package name?
Epoch Tag, see http://www.rpm.org/max-rpm-snapshot/s1-rpm-inside-tags.html
jlc
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
However, when running nmap on them, one betrays the port that SSH is running
on, and the other does not.
What does betray mean?
I have shut down iptables on both machines and the behaviour remains this way.
What could be the cause?
Public facing machines w/ iptables off?
Specifically, how
I understand what it is and how it works but the why has eluded me.
You need to do maintenance on hardware and you don't want it running
wasting power or it happen to be off such as the case with many client boxes?
You need to start a node and you don't want it running wasting power while
you
The more people test and provide feedback, the
quicker we can move it out of testing, into the elrepo repository.
Dag,
I got these on my cluster at work, it will be exercised thoroughly this
weekend with a tb of data changing, I'll report back next week.
jlc
/snip
just noticed it still has the old 2.6.18-128 kernel instead of the new
2.6.18-194.17. What gives?
/etc/grub.conf points at 2.6.18-194.17, but when I reboot, 2.6.18-128 comes up.
Any suggestions?
/snip
kernel /vmlinuz-2.6.18-194.17.1.el5 ro root=/dev/md1
Yup, looking at your
That doesn't really solve the issue, it only stops you from getting
important kernel updates.
An unfortunate way of life if you insist on using HP's modules. They do
their own QA testing and it takes a moons age. I have never had a need
to use the latest cciss for example versus the shipped
ehh -- The version in RawHide will build and run under C5,
with some tedious gathering and packaging of buildchain work
[herr...@centos-5 zoneminder]$ rpm -q zoneminder
zoneminder-1.24.2-5orc
[herr...@centos-5 zoneminder]$ ls *src.rpm | grep orc | wc
28 281021
[herr...@centos-5
I then edit the openssl.spec file, and add 2 lines to that in the appropriate
place:
Patch88: openssl-ssl-h.patch
And
%patch88 -p1
/snip
How do I go about troubleshooting this?
Post the top lines in your patch file, are you certain p1 is what you need?
My bet is that it can't find the
I am wondering if I need to worry about stripe and width though as
mine resides on a logical volume residing on a hardware-controlled
RAID 6 device (i.e., one slice as far as the OS is concerned).
25 TB on a single volume, not distributed? Huh, let me know how long
that takes to check the first
Is there a way to copy it to a hardware RAID 1 array of two
drives in a similiar way?
The same way? The underlying OS knows nothing of the physical geometry
behind a hardware raid controller. Find out what dev it is, sd? and do what
you did, so long as it's exactly the same size or bigger.
Would going from a single 500G sata to two 500G sata drives in
hardware raid 1 work? Just wandering if some space will be lost using
raid 1?
Again like I said: The same or bigger. Often HW raid controllers coerce the
size of a drive down a gig or so slightly different drives can all work
I been working on for about a month and half on setup drbd and nfs. I keep
running
into issue with the way heartbeat/pacemaker handles nfs. Does anyone know a
good
way to set up a HA NFS server with DRBD and Heartbeat and NFS. I am willing
to share
my pain in setting it up.
It involves a bit
/snip
Allowing Access:
/snip
Out of curiosity, when you read the log, did you attempt the suggestion w/o
success?
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
I have a few dvr's that I am contemplating converting to Zoneminder (don't
know any other app) and would prefer to use CentOS over fedora obviously
but the only rpm I see exists for Fedora.
I don't want the hassle of manually compiling this, as we have our own config
mgmt.
Anyone using this rpm
We gave up on zoneminder and moved to motion, easy to build RPM's and
configure.
http://www.lavrsen.dk/twiki/bin/view/Motion/WebHome
Looks nice, I'll sub to that list but quick last question: After a read of the
docs, I
get the idea it's not designed to output a continuous mpeg for each camera
how can i exclude certain mirrors from yum in centos 5?
http://lmgtfy.com/?q=yum+exclude+mirrorsl=1
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
I am running CentOS 5.5 in command line only. Reason being I am standing up
an IDS system on it.
I have installed the NX client/node and server pkgs. I have installed the NX
client for windows on a Vista box. When I attemtp to connect from the Vista
to CentOS, I get this:
Huh? So no
I replaced the sshd_config on the affected server with a copy of
that which came with the package and the same error occurs. Whatever
is causing this, it does not seem to be related the the sshd_config
file.
Selinux enabled?
___
CentOS mailing list
Selinux enabled?
Yes.
Any errors when you start sshd? Check the contexts...
Maybe a fixfiles -R on that package.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
Set selinux=permissive in /etc/selinux/config
Rebooted system
tried to restart sshd
Starting sshd: Auto configuration failed
3600:error:0E065068:configuration file routines:STR_COPY:variable
has no value:conf_def.c:629:line 207
[FAILED]
Same error. I point out at that the other servers that
My reboot times are regular, (still on F12 on this
machine) but I always copy the kernel files into a subdir 'tmp-backups' so I
can get them back if needed, even if yum deletes them.
Huh, ok... What do you do with *just* the kernel?
Let me know how that works if you ever want to boot from it?
But if I try to make the installation in a RAID-1 with a single disk,
the system does not allow it. There is any way to do it?
Drop into a shell, create (force) the md devices, partition etc then
drop into the installer and go...
___
CentOS mailing list
Stop it at the Edge Router not the machine.
Fair enough, but now I have to manually scour the logs and
maintain a dynamic block list?
Adding layers of security become problems like you are getting.
I agree, and if my edge router had the functionality to inspect
http requests I would:)
Ban the
http://www.fail2ban.org/wiki/index.php/Fail2ban:Community_Portal
Question about persistant IP bans over restart
I think you need to adapt the example to CentOS/RH
Yeah, I saw that one and implemented it. I think I have to rewrite
the action scripts my jails use. The odd part is the initial
Or block all networks like china,japan,india and so on. Can get these from
ICANN.
Actually. that might just be enough, I know this site won't need access
from other that NA addresses which is an easy rule to build permanently.
Thanks,
jlc
___
CentOS
I created a filter and verified it with fail2ban-regex against
actual lines in my log and it works. During restarts of fail2ban,
only some previous ip's get banned immediately whereas some need a
reoccurrence despite the jail's config specification of maxretry and
findtime suggesting the entries
101 - 200 of 972 matches
Mail list logo
