Discuss March 2007

discuss@lists.centos.org

337 participants
401 discussions

net.ipv4 TCP/IP Optimizations = sysctl.conf?
by Karl R. Balsmeier 19 Mar '07

19 Mar '07

If I execute these via command line, will they persist after a reboot? Or, should I be putting these into a file like /etc/sysctl.conf? --------------snip-------------- /sbin/sysctl -w net.ipv4.tcp_max_syn_backlog=2048 /sbin/sysctl -w net.ipv4.tcp_fin_timeout=30 /sbin/sysctl -w net.ipv4.tcp_keepalive_intvl=10 /sbin/sysctl -w net.ipv4.tcp_keepalive_probes=7 /sbin/sysctl -w net.ipv4.tcp_keepalive_time=1800 /sbin/sysctl -w net.ipv4.tcp_max_tw_buckets=360000 /sbin/sysctl -w net.ipv4.tcp_synack_retries=3 /sbin/sysctl -w net.ipv4.tcp_rmem="4096 87380 16777216" /sbin/sysctl -w net.ipv4.tcp_wmem="4096 87380 16777216" /sbin/sysctl -w net.ipv4.tcp_mem="8388608 8388608 8388608" --------------snip--------------- p.s. these are meant for a specific technology we use, so not sure if everyone reading would be best served by using them, mileage may vary! : / -karlski

5 6

sysctl errors
by Karl R. Balsmeier 19 Mar '07

19 Mar '07

I get these errors from the following settings in /etc/sysctl.conf file: # Custom Settings: net.ipv4.tcp_max_syn_backlog=2048 net.ipv4.tcp_fin_timeout=30 net.ipv4.tcp_keepalive_intvl=10 net.ipv4.tcp_keepalive_probes=7 net.ipv4.tcp_keepalive_time=1800 net.ipv4.tcp_max_tw_buckets=360000 net.ipv4.tcp_synack_retries=3 net.ipv4.tcp_rmem="4096 87380 16777216" net.ipv4.tcp_wmem="4096 87380 16777216" net.ipv4.tcp_mem="8388608 8388608 8388608" ---------errors----------- # sysctl -p [errors] error: unknown error 22 setting key 'net.ipv4.tcp_rmem' error: unknown error 22 setting key 'net.ipv4.tcp_wmem' error: unknown error 22 setting key 'net.ipv4.tcp_mem' any advice?

2 1

system-display-config doesn't run right
by Mark Hull-Richter 19 Mar '07

19 Mar '07

This might be a gnome problem, but.... I've noticed that, on occasion (frequently), when I try to look at or modify my display configuration, the Applications->System Settings->Display doesn't come up. It asks for the root password, which I give it, and then nothing. I ran system-config-display form the command line, but it dies thus: [root@mhullrichter mhr]# system-config-display * no cardMem, assuming 8192 * no cardMem, assuming 8192 Trying with card: VESA driver (generic) * no cardMem, assuming 8192 Traceback (most recent call last): File "/usr/share/system-config-display/xconf.py", line 380, in ? dialog = xConfigDialog.XConfigDialog(hardware_state, xconfig, rhpl.videocard.VideoCardInfo()) File "/usr/share/system-config-display/xConfigDialog.py", line 566, in __init__ self.xml.get_widget("secondMonitorLabel").set_text(monitor_list[1].model name) TypeError: GtkLabel.set_text() argument 1 must be string, not None I seem to be missing an argument from the call, but I'm not sure what goes there, and even so, when I run it from the menu, shouldn't it have the right data and run? Thanks. Mark Hull-Richter Linux Kernel Engineer (949) 680-3082 - Office (949) 680-3001 - Fax (949) 632-8403 - Mobile mhull-richter(a)datallegro.com <mailto:mhull-richter@datallegro.com> www.datallegro.com <http://www.datallegro.com> 85 Enterprise, 2nd Floor, Aliso Viejo, CA 92656

3 3

installing perl module in centos/rpmforge vs. cpan methods
by Karl R. Balsmeier 19 Mar '07

19 Mar '07

Hi, I just used RPMforge package perl-Mail-Sender, and am looking at the perl cpan site for info, anyone got a simple way to determine if the module is working properly? I am trying to use the cpan info, but it's a bit cryptic (as perl is wont to be). -it says to use this: #!/usr/local/bin/perl use ExtUtils::Installed; my $instmod = ExtUtils::Installed->new(); foreach my $module ($instmod->modules()) { my $version = $instmod->version($module) || "???"; print "$module -- $version\n"; } but all I get is: Perl -- 5.8.5 which seems to indicate doing a wget http://dag.wieers.com/rpm/packages/perl-Mail-Sender/perl-Mail-Sender-0.8.13… rpm -ivh perl-Mail-Sender-0.8.13-1.el4.rf.noarch.rpm is either working and thie script above just cant see it, or the module is installed and requires some follow up. Both of these are just theoretical. Any input from the wise listers other than "read all of cpan since you have a million hours free time". perl -MCPAN -e 'install Mail::Sender' simply hangs on the FTP transaction, thus the reason I did this from RPMforge instead. -krb

2 3

Changing from NTFS to ext3
by Mark Hull-Richter 19 Mar '07

19 Mar '07

I want to convert all my NTFS partitions to ext3 since I'm no longer running Windows, at least on my main machine at home. I'm planning to tar the NTFS partitions to an existing ext3, reformat the partition as ext3 and untar it back. What is the best way to accomplish step two of this - reformatting the partition? I've only done this so far during the installation phase, and that was for partitions I knew I could destroy with impunity. I'm guessing I unmount the drive, <do something magic here> and remount it. It's the magic part I don't know. Thanks. Mark Hull-Richter Linux Kernel Engineer (949) 680-3082 - Office (949) 680-3001 - Fax (949) 632-8403 - Mobile mhull-richter(a)datallegro.com <mailto:mhull-richter@datallegro.com> www.datallegro.com <http://www.datallegro.com> 85 Enterprise, Second Floor, Aliso Viejo, CA 92656

5 12

Fun with CentOS and Windows
by Mark Hull-Richter 19 Mar '07

19 Mar '07

I installed the memory and the CPU and the motherboard and the new video card all together tonight. CentOS installed very nicely and I'm currently running the CentOS-Plus version, with which I _can_ access my Windows partitions. Well, one of the disks is getting flaky and I'm having trouble with it - of course, that's the one I need the most. Glitches, with which I would be delighted to obtain feedback and/or advice: 1) In the boot screen, it tells me "CPU0 Memory information: single channel, 64-bits." I have 2Gb of DDR2 PC6400 (800MHz) memory - is this a problem, or does it mean something else entirely? (ECS NFORCE4M-A m/b, Phoenix BIOS 6.00PG) 2) While the system is booting up, it tells me: MP-BUG: 8254 timer is not connected to APIC What does this mean and is there a way around it? 3) CentOS does not appear to have the drivers for my video card or monitor. I have an e-GEForce 7100gs card and an Envision en910e monitor (yeah, that old 19" CRT that has been working fine for 3+ years). How can I get more than 800x600 (need 1280x1024 to work properly? And please don't say buy a new monitor - no money for that for a while.... 4) Here's the bad part (sort of) - I can't boot my Windows any more. It comes part way up and reboots, whether I try to run Safe Mode (hahahahahaha) or Crashing (I mean Normal) Mode. The video card is different, and I thought that might be the problem, but it should come up in safe mode no matter what - no video driver loaded. (I also went from a P4 to an Athlon 64 X2 - could that be part/all of it?). Some of this might be O/T, so please point me at the right place if so (except that last one - there is no right place for that stuff....)-; Thanks. mhr (aka mhull-richter(a)datallegro.com) -- [This message is sent in strict accordance with both Google and Yahoo Terms of Service. If you are receiving this second-hand, this sender disclaims all responsibility for your response.]

10 13

4.4 on ESX 2.5.4
by Dhivan Djaganu 19 Mar '07

19 Mar '07

hi All, I followed this and installed 4.4 on ESX 2.5.4 without a hitch. Any pitfalls I shall be aware of before I try yup update? TIA

1 2

What's the best way to convert a whole set of file systems?
by Mark Hull-Richter 19 Mar '07

19 Mar '07

I am currently running a Windows XP system at home with around 100+ Mb in use over ~400Mb of NTFS file systems. I am installing CentOS 4.4 on it when I change out the mobo/cpu/mem/video combo I just bought. I want to convert all the file systems to (probably) Reiserfs or maybe ext3, but I need to do them one at a time because I only have enough transfer space to accommodate the largest one, or at least that's my belief. That would mean at least two copies per partition converted, and I have six partitions to convert, from ~14Gb to over 85Gb (in one, only - the rest are 30Gb or smaller). 1) Is there a good way to do whole fs conversions, specifically from NTFS to reiserfs or ext3? 2) Do I even need to do this (i.e., do any of the CentOS/Linux kernels support read AND write to NTFS)? 3) Is there, by any chance, and in-place converter from NTFS to any Linux fs, preferably reiserfs or ext3? Also, the last time I installed CentOS on a system (I've done about six or seven so far) I don't remember seeing reiserfs as one of the supported fs's for configuring during the installation process - am I blind or is this really the case? I like reiserfs primarily because it is really good with many small files, and I have tons of them - around 100k files under 10k. Thanks. mhr

11 26

Upgrading RH EL by CentOS
by Joachim Backes 19 Mar '07

19 Mar '07

Hi, is it possible to upgrade a RH EL system with version x to a CentOS system with a version > x? All comments are welcome. Regards Joachim Backes <joachim.backes(a)rhrk.uni-kl.de> University of Kaiserslautern,Computer Center [RHRK], Systems and Operations, High Performance Computing, D-67653 Kaiserslautern, PO Box 3049, Germany -------------------------------------------------- Phone: +49-631-205-2438, FAX: +49-631-205-3056 http://hlrwm.rhrk.uni-kl.de/home/staff/backes.html

3 2

Configuration Help with HA/VirtualIP
by Todd Reed 19 Mar '07

19 Mar '07

I've been bashing my head into a wall the past few days on this. I am trying to setup a Load Balanced MySQL Cluster. I can get the MySQL NDB Clusters working, but when I try to install the HA Load Balancers, I can't get past the step of the shared Virtual IP. I'm basing my sequence on HowToForge Guide written for Debain (http://www.howtoforge.com/loadbalanced_mysql_cluster_debian_p7). I've added my own tweaks and use the appropriate CentOS x64 commands and configurations. I am creating a loopback interface with the following configs: vi /etc/sysconfig/network-scripts/ifcfg-lo:0 DEVICE=lo:0 IPADDR=10.98.51.10 NETMASK=255.255.254.0 NETWORK=10.98.50.0 # If you're having problems with gated making 127.0.0.0/8 a martian, # you can change this to something else (255.255.255.255, for example) BROADCAST=10.98.51.255 ONBOOT=yes NAME=loopback The Guide shows this for Debian vi /etc/network/interfaces auto lo:0 iface lo:0 inet static address 192.168.0.105 netmask 255.255.255.255 pre-up sysctl -p > /dev/null I think I am almost there, but just can't place my figure on my error. Can anyone help guide me?

1 0

← Newer
1
...
18
19
20
21
22
23
24
...
41
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Discuss March 2007