Wednesday, September 10, 2008
Application failed to initialize 0xc0000135 Google Chrome known issue
Thursday, January 31, 2008
AIX commands you should not leave home without
As you know, AIX® has a vast array of commands that enable you to do a multitude of tasks. Depending on what you need to accomplish, you use only a certain subset of these commands. These subsets differ from user to user and from need to need. However, there are a few core commands that you commonly use. You need these commands either to answer your own questions or to provide answers to the queries of the support professionals.
In this article, I'll discuss some of these core commands. The intent is to provide a list that you can use as a ready reference. While the behavior of these commands should be identical in all releases of AIX, they have been only tested under AIX 5.3.
Note:
The bootinfo command discussed in the following paragraphs is NOT a user-level command and is NOT supported in AIX 4.2 or later.
How would I know if I am running a 32-bit kernel or 64-bit kernel?
To display if the kernel is 32-bit enabled or 64-bit enabled, type:
bootinfo -K |
How do I know if I am running a uniprocessor kernel or a multiprocessor kernel?
/unix is a symbolic link to the booted kernel. To find out what kernel mode is running, enter ls -l /unix and see what file /unix it links to. The following are the three possible outputs from the ls -l /unix command and their corresponding kernels:
/unix -> /usr/lib/boot/unix_up # 32 bit uniprocessor kernel |
Note:
AIX 5L Version 5.3 does not support a uniprocessor kernel.
How can I change from one kernel mode to another?
During the installation process, one of the kernels, appropriate for the AIX version and the hardware in operation, is enabled by default. Let us use the method from the previous question and assume the 32-bit kernel is enabled. Let us also assume that you want to boot it up in the 64-bit kernel mode. This can be done by executing the following commands in sequence:
ln -sf /usr/lib/boot/unix_64 /unix |
The /dev/hdiskxx directory is where the boot logical volume /dev/hd5 is located. To find out what xx is in hdiskxx, run the following command:
lslv -m hd5 |
Note:
In AIX 5.2, the 32-bit kernel is installed by default. In AIX 5.3, the 64-bit kernel is installed on 64-bit hardware and the 32-bit kernel is installed on 32-bit hardware by default.
How would I know if my machine is capable of running AIX 5L Version 5.3?
AIX 5L Version 5.3 runs on all currently supported CHRP (Common Hardware Reference Platform)-based POWER hardware.
How would I know if my machine is CHRP-based?
Run the prtconf command. If it's a CHRP machine, the string chrp appears on the Model Architecture line.
How would I know if my System p machine (hardware) is 32-bit or 64-bit?
To display if the hardware is 32-bit or 64-bit, type:
bootinfo -y |
How much real memory does my machine have?
To display real memory in kilobytes (KB), type one of the following:
bootinfo -r |
lsattr -El sys0 -a realmem |
Can my machine run the 64-bit kernel?
64-bit hardware is required to run the 64-bit kernel.
What are the values of attributes for devices in my system?
To list the current values of the attributes for the tape device, rmt0, type:
lsattr -l rmt0 -E |
To list the default values of the attributes for the tape device, rmt0, type:
lsattr -l rmt0 -D |
To list the possible values of the login attribute for the TTY device, tty0, type:
lsattr -l tty0 -a login -R |
To display system level attributes, type:
lsattr -E -l sys0 |
How many processors does my system have?
To display the number of processors on your system, type:
lscfg | grep proc |
How many hard disks does my system have and which ones are in use?
To display the number of hard disks on your system, type:
lspv |
How do I list information about a specific physical volume?
To find details about hdisk1, for example, run the following command:
lspv hdisk1 |
How do I get a detailed configuration of my system?
Type the following:
lscfg |
The following options provide specific information:
-p | Displays platform-specific device information. The flag is applicable to AIX 4.2.1 or later. |
-v | Displays the VPD (Vital Product Database) found in the customized VPD object class. |
For example, to display details about the tape drive, rmt0, type:
lscfg -vl rmt0 |
You can obtain very similar information by running the prtconf command.
How do I find out the chip type, system name, node name, model number, and so forth?
The uname command provides details about your system.
uname -p | Displays the chip type of the system. For example, PowerPC. |
uname -r | Displays the release number of the operating system. |
uname -s | Displays the system name. For example, AIX. |
uname -n | Displays the name of the node. |
uname -a | Displays the system name, nodename, version, machine ID. |
uname -M | Displays the system model name. For example, IBM, 9114-275. |
uname -v | Displays the operating system version. |
uname -m | Displays the machine ID number of the hardware running the system. |
uname -u | Displays the system ID number. |
What version, release, and maintenance level of AIX is running on my system?
Type one of the following:
oslevel -r |
lslpp -h bos.rte |
How can I determine which fileset updates are missing from a particular AIX level?
To determine which fileset updates are missing from 5300-04, for example, run the following command:
oslevel -rl 5300-04 |
What SP (Service Pack) is installed on my system?
To see which SP is currently installed on the system, run the oslevel -s command. Sample output for an AIX 5L Version 5.3 system, with TL4, and SP2 installed would be:
oslevel –s |
Is a CSP (Concluding Service Pack) installed on my system?
To see if a CSP is currently installed on the system, run the oslevel -s command. Sample output for an AIX 5L Version 5.3 system, with TL3, and CSP installed would be:
oslevel –s |
How do I create a file system?
The following command will create, within volume group testvg, a jfs file system of 10MB with mounting point /fs1:
crfs -v jfs -g testvg -a size=10M -m /fs1 |
The following command will create, within volume group testvg, a jfs2 file system of 10MB with mounting point /fs2 and having read only permissions:
crfs -v jfs2 -g testvg -a size=10M -p ro -m /fs2 |
How do I change the size of a file system?
To increase the /usr file system size by 1000000 512-byte blocks, type:
chfs -a size=+1000000 /usr |
Note:
In AIX 5.3, the size of a JFS2 file system can be shrunk as well.
How do I mount a CD?
Type the following:
mount -V cdrfs -o ro /dev/cd0 /cdrom |
How do I mount a file system?
The following command will mount file system /dev/fslv02 on the /test directory:
mount /dev/fslv02 /test |
How do I mount all default file systems (all standard file systems in the /etc/filesystems file marked by the mount=true attribute)?
The following command will mount all such file systems:
mount {-a|all} |
How do I unmount a file system?
Type the following command to unmount /test file system:
umount /test |
How do I display mounted file systems?
Type the following command to display information about all currently mounted file systems:
mount |
How do I remove a file system?
Type the following command to remove the /test file system:
rmfs /test |
How can I defragment a file system?
The defragfs command can be used to improve or report the status of contiguous space within a file system. For example, to defragment the file system /home, use the following command:
defragfs /home |
Which fileset contains a particular binary?
To show bos.acct contains /usr/bin/vmstat, type:
lslpp -w /usr/bin/vmstat |
Or to show bos.perf.tools contains /usr/bin/svmon, type:
which_fileset svmon |
How do I display information about installed filesets on my system?
Type the following:
lslpp -l |
How do I determine if all filesets of maintenance levels are installed on my system?
Type the following:
instfix -i | grep ML |
How do I determine if a fix is installed on my system?
To determine if IY24043 is installed, type:
instfix -ik IY24043 |
How do I install an individual fix by APAR?
To install APAR IY73748 from /dev/cd0, for example, enter the command:
instfix -k IY73748 -d /dev/cd0 |
How do I verify if filesets have required prerequisites and are completely installed?
To show which filesets need to be installed or corrected, type:
lppchk -v |
How do I get a dump of the header of the loader section and the symbol entries in symbolic representation?
Type the following:
dump -Htv |
How do I determine the amount of paging space allocated and in use?
Type the following:
lsps -a |
How do I increase a paging space?
You can use the chps -s command to dynamically increase the size of a paging space. For example, if you want to increase the size of hd6 with 3 logical partitions, you issue the following command:
chps -s 3 hd6 |
How do I reduce a paging space?
You can use the chps -d command to dynamically reduce the size of a paging space. For example, if you want to decrease the size of hd6 with four logical partitions, you issue the following command:
chps -d 4 hd6 |
How would I know if my system is capable of using Simultaneous Multi-threading (SMT)?
Your system is capable of SMT if it's a POWER5-based system running AIX 5L Version 5.3.
How would I know if SMT is enabled for my system?
If you run the smtctl command without any options, it tells you if it's enabled or not.
Is SMT supported for the 32-bit kernel?
Yes, SMT is supported for both 32-bit and 64-bit kernel.
How do I enable or disable SMT?
You can enable or disable SMT by running the smtctl command. The following is the syntax:
smtctl [ -m off | on [ -w boot | now]] |
The following options are available:
-m off | Sets SMT mode to disabled. |
-m on | Sets SMT mode to enabled. |
-w boot | Makes the SMT mode change effective on next and subsequent reboots if you run the bosboot command before the next system reboot. |
-w now | Makes the SMT mode change immediately but will not persist across reboot. |
If neither the -w boot or the -w now options are specified, then the mode change is made immediately. It persists across subsequent reboots if you run the bosboot command before the next system reboot.
How do I get partition-specific information and statistics?
The lparstat command provides a report of partition information and utilization statistics. This command also provides a display of Hypervisor information.
Volume groups and logical volumes
How do I know if my volume group is normal, big, or scalable?
Run the lsvg command on the volume group and look at the value for MAX PVs. The value is 32 for normal, 128 for big, and 1024 for scalable volume group.
How to create a volume group?
Use the following command, where s partition_size sets the number of megabytes (MB) in each physical partition where the partition_size is expressed in units of MB from 1 through 1024. (It's 1 through 131072 for AIX 5.3.) The partition_size variable must be equal to a power of 2 (for example: 1, 2, 4, 8). The default value for standard and big volume groups is the lowest value to remain within the limitation of 1016 physical partitions per physical volume. The default value for scalable volume groups is the lowest value to accommodate 2040 physical partitions per physical volume.
mkvg -y name_of_volume_group -s partition_size list_of_hard_disks |
How can I change the characteristics of a volume group?
You use the following command to change the characteristics of a volume group:
chvg |
How do I create a logical volume?
Type the following:
mklv -y name_of_logical_volume name_of_volume_group number_of_partition |
How do I increase the size of a logical volume?
To increase the size of the logical volume represented by the lv05 directory by three logical partitions, for example, type:
extendlv lv05 3 |
How do I display all logical volumes that are part of a volume group (for example, rootvg)?
You can display all logical volumes that are part of rootvg by typing the following command:
lsvg -l rootvg |
How do I list information about logical volumes?
Run the following command to display information about the logical volume lv1:
lslv lv1 |
How do I remove a logical volume?
You can remove the logical volume lv7 by running the following command:
rmlv lv7 |
The rmlv command removes only the logical volume, but does not remove other entities, such as file systems or paging spaces that were using the logical volume.
How do I mirror a logical volume?
- mklvcopy LogicalVolumeName Numberofcopies
- syncvg VolumeGroupName
How do I remove a copy of a logical volume?
You can use the rmlvcopy command to remove copies of logical partitions of a logical volume. To reduce the number of copies of each logical partition belonging to logical volume testlv, enter:
rmlvcopy testlv 2 |
Each logical partition in the logical volume now has at most two physical partitions.
Queries about volume groups
To show volume groups in the system, type:
lsvg |
To show all the characteristics of rootvg, type:
lsvg rootvg |
To show disks used by rootvg, type:
lsvg -p rootvg |
How to add a disk to a volume group?
Type the following:
extendvg VolumeGroupName hdisk0 hdisk1 ... hdiskn |
How do I find out what the maximum supported logical track group (LTG) size of my hard disk?
You can use the lquerypv command with the -M flag. The output gives the LTG size in KB. For instance, the LTG size for hdisk0 in the following example is 256 KB.
/usr/sbin/lquerypv -M hdisk0 |
You can also run the lspv command on the hard disk and look at the value for MAX REQUEST.
What does syncvg command do?
The syncvg command is used to synchronize stale physical partitions. It accepts names of logical volumes, physical volumes, or volume groups as parameters.
For example, to synchronize the physical partitions located on physical volumes hdisk6 and hdisk7, use:
syncvg -p hdisk4 hdisk5 |
To synchronize all physical partitions from volume group testvg, use:
syncvg -v testvg |
How do I replace a disk?
-
extendvgVolumeGroupName hdisk_new -
migratepvhdisk_bad hdisk_new -
reducevg -dVolumeGroupName hdisk_bad
How can I clone (make a copy of ) the rootvg?
You can run the alt_disk_copy command to copy the current rootvg to an alternate disk. The following example shows how to clone the rootvg to hdisk1.
alt_disk_copy -d hdisk1 |
How can I display or set values for network parameters?
The no command sets or displays current or next boot values for network tuning parameters.
How do I get the IP address of my machine?
Type one of the following:
ifconfig -a |
For example, type host cyclop.austin.ibm.com.
How do I identify the network interfaces on my server?
Either of the following two commands will display the network interfaces:
lsdev -Cc if |
ifconfig -a |
To get information about one specific network interface, for example, tr0, run the command:
ifconfig tr0 |
How do I activate a network interface?
To activate the network interface tr0, run the command:
ifconfig tr0 up |
How do I deactivate a network interface?
For example, to deactivate the network interface tr0, run the command:
ifconfig tr0 down
Thursday, December 13, 2007
EFFICIENT COMMANDS
Better: grep something somefile
Why: You're running one program (grep) instead of two (cat and grep).
Bad: ps -ef | grep something | grep -v grep
Better: ps -ef | grep [s]omething
Why: You're running two commands (grep) instead of three (ps
and two greps).
Bad: cat /dev/null > somefile
Better: > somefile
Why: You're running a command (cat) with I/O redirection,
instead of just redirection.
Although the bad way will have the
same result, the good way is far
faster. This may seem trivial, but
the benefits will really show when
dealing with large files or loops.
Thursday, July 19, 2007
Setting Up PureFTP server
1. download pureftpd website http://www.pureftpd.org/project/pure-ftpd/doc
http://download.pureftpd.org/pub/pure-ftpd/releases/pure-ftpd-1.0.21.tar.gz
2. compile i suggest you use "./configure --with-everything" to install all features
3. this ftp server have versatile auth methods (read the documention). but in my case i've use ftp authentiacation based on a independent password file.. below are the
steps.
a. create an "ftpgroup" group and an "ftpuser" user.
groupadd ftpgroup
useradd -g ftpgroup -d /dev/null -s /etc ftpuser
b. execute the following command to add the user
pure-pw useradd testaccount -u ftpuser -d /home/ftpusers/testaccount -N 200
where
user - testaccount
uid - ftpuser
chroot directory - /home/ftpusers/testaccount
disk quota - 200 MB
* this will create "/etc/pureftpd.passwd" by default
c. execute "pure-pw mkdb" to save the changes
* this will create "/etc/pureftpd.pdb"
* remember always execute this after changes made to password file
d. to check account info execute the following
pure-pw show testaccount
OUTPUT:
Login : testaccount
Password : $1$G5jGYrk0$n8vs/BsYv5a2cypCLMStm.
UID : 508 (ftpuser)
GID : 509 (ftpgroup)
Directory : /home/ftpusers/testaccount/./
Full name :
Download bandwidth : 0 Kb (unlimited)
Upload bandwidth : 0 Kb (unlimited)
Max files : 0 (unlimited)
Max size : 200 Mb (enabled)
Ratio : 0:0 (unlimited:unlimited)
Allowed local IPs :
Denied local IPs :
Allowed client IPs :
Denied client IPs :
Time restrictions : 0000-0000 (unlimited)
Max sim sessions : 0 (unlimited)
4. run pureftp server
/usr/local/sbin/pure-ftpd -j -lpuredb:/etc/pureftpd.pdb &
Documentation:
http://www.pureftpd.org/project/pure-ftpd/doc
Sunday, May 06, 2007
LVM TUTORIAL BASICS
LVM TUTORIAL BASICS V 1.0 03/21/07
by derrick caluag
[root@intel-4 ~]# fdisk -l
Disk /dev/sda: 145.4 GB, 145492017152 bytes
255 heads, 63 sectors/track, 17688 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Device Boot Start End Blocks Id System
/dev/sda1 * 1 16 128488+ 83 Linux
/dev/sda2 17 1060 8385930 83 Linux
/dev/sda3 1061 1321 2096482+ 82 Linux swap
/dev/sda4 1322 17688 131467927+ 8e Linux LVM
INITIALIZATION
-- Create Physical Volume
pvcreate /dev/sda4
-- Create Volume Group
vgcreate VolG00 /dev/sda4
-- Create Logical Group (this will create 21G for LVhome)
lvcreate -L21000 VolG00 -n LVhome
-- Format the create Group with desired file system
mkfs.ext3 -b 4096 /dev/VolG00 / LVhome
-- Create Mount point on fstab
vi /etc/fstab
/dev/VolG00/LVhome /home ext3 defaults 1 2
-- Backup files on mountpoint
tar –zcvf /tmp/home.tgz /home
mv /home home2
mkdir home
-- Mount the partition
mount /dev/VolG00/LVhome
-- Untar the backup file back to the mount point
cd /home
cp /tmp/home.tgz
tar –zxvf home.tgz
EXTENDING LVM
-- Unmount partition
umount /dev/VolG00/LVhome
Extend
lvextend -L+1G /dev/VolG00/LVhome
add 1G to current size
or
lvextend -L12G /dev/VolG00/LVhome
extend size to 12G
-- Check extended
e2fsck -f /dev/VolG00/LVhome
-- Resize the
resize2fs /dev/VolG00/LVhome
-- Remount the partition
mount /dev/VolG00/LVhome
Thursday, March 15, 2007
Installing Perl Modules
CPAN
The Comprehensive Perl Archive Network (CPAN) holds many Perl modules and scripts, distributed among a large number of mirrors. The CPAN module provides an interface to query and install modules from CPAN. These notes document the setup and usage of CPAN, with attention to common tasks and pitfalls along the way.
Any Perl modules used by a site should be archived locally. Doing so ensures the modules can be installed on new systems, even if removed from CPAN.
Use CPAN::Reporter to submit module test results. Try SmokeAuto to automate module testing.
Alternatives
Other means of installing perl modules from CPAN include the following, which are not discussed in detail here.
- CPANPLUS. “CPAN++ (also referred to and pronounced as CPANPLUS) is a new and flexible method of perl module management and installation using the CPAN. It aims to be a rewrite, and in time a replacement to the current CPAN module. In addition to fixing some long-standing problems, CPAN++ includes new features, such as module uninstall.”
- PAR - Cross-Platform Packaging and Deployment tool.
- Perl Package Manager (PPM) - Included with ActiveState Perl.
- Other. Some vendors make perl modules available via other methods, such as ActiveState’s PPM, the FreeBSD ports system, or similar package software. These offer tight integration with the system in question, but may include outdated modules or use different naming conventions and filesystem layouts.
For example, p5-libwww is the FreeBSD port of the module distribution libwww-perl on CPAN, and LWP::UserAgent a module found in that distribution. The following commands are equivalent means to install libwww-perl via the FreeBSD ports system, yum on RedHat systems, or the cpan command included with modern versions of CPAN.
# portinstall p5-libwww
# yum -y install perl-libwww-perl
# cpan -i LWP::UserAgent
To lookup the documentation for a module, use http://search.cpan.org/perldoc/Module::Name. The documentation should also show the distribution a module belongs to, for example Data::Dumper. Other search interfaces include the CPAN Search Site or the CPAN shell search features.
- Conversion scripts. cpan2rpm helps convert Perl modules into RedHat Package Manager (RPM) files. The resulting *.rpm packages can be installed via a local Yellowdog Updater, Modified (YUM) server. On Debian, try dh-make-perl --install --cpan Module::Name.
perl Upgrade Bug
The first time CPAN is invoked, it will run through a set of questions, such as which CPAN mirror site to use. Following the questions, CPAN may recommend that you install Bundle::CPAN or Bundle::libnet. Avoid installing these bundles!
Older versions of CPAN have a bug whereby a new version of perl itself may be installed. To avoid this bug, install the latest version of CPAN first. Setting the FTP_PASSIVE=1 environment variable avoids another common problem where active File Transfer Protocol (FTP) connections are blocked by a local firewall.
# perl -MCPAN -e '$ENV{FTP_PASSIVE} = 1; install CPAN'
If already in the CPAN shell following the preferences questions, use:
cpan> install CPAN
For reference, the bug is that Bundle::CPAN or Bundle::libnet may list Data::Dumper as a dependency. Data::Dumper is a built-in perl module, so when old versions of CPAN lookup Data::Dumper, they find a new version of perl to install. This bug has been fixed in current versions of CPAN.
Active FTP versus Firewalls
Network problems may prevent access to CPAN mirror sites. The chief problem will be utilities set to use active FTP, which most firewalls block. Persue Active FTP vs. Passive FTP, a Definitive Explanation for more details. I recommend enabling passive FTP by default, though this setting will need to be done for a variety of utilities CPAN might run, such as wget or Net::FTP. Alternatives include disabling the client firewall to allow active FTP connections back from the FTP server, or configuring a local FTP proxy.
Set the FTP_PASSIVE environment variable before running cpan.
# env FTP_PASSIVE=1 cpan -i Net::FTP
During the install of the libnet distribution, enable passive FTP.
…
Ah, I see you already have installed libnet before.
Do you want to modify/update your configuration (y|n) ? [no] y
…
Should all FTP connections be passive (y|n) ? [no] y
If Net::FTP is already installed, edit the libnet.cfg configuration file to use passive FTP.
$ perl -le 'for (@INC) { $_ .= "/".$ARGV[0]; print if -f }' Net/libnet.cfg
/usr/local/lib/perl5/5.8.1/Net/libnet.cfg
$ grep passive /usr/local/lib/perl5/5.8.1/Net/libnet.cfg
'ftp_int_passive' => 1,
If Net::FTP does not work, or certain utilities take too long to time out, consider the following options.
- Disable Net::FTP.
If Net::FTP does not work, try adding the following line to the CPAN/Config.pm preferences file. More information on dontload_hash posted to the perl.perl5.porters newsgroup. Also consider installing LWP::UserAgent, which is more capable than Net::FTP.
'dontload_hash' => { 'Net::FTP' => 1 },
- Disable utilities that do not work
To reconfigure CPAN to not call certain utilities, replace the path to the utility with a single space.
$ grep ncftp ~/.cpan/CPAN/MyConfig.pm
'ncftp' => q[ ],
'ncftpget' => q[ ],
Preferences Questions
CPAN will ask for preferences settings any time the Config.pm preferences file does not exist or is out of date. The default choice should be acceptable for most.
- CPAN build and cache directory? [/root/.cpan]
I use /var/spool/cpan, as the root filesystem tends to be small on my systems, and the spool area makes more sense to me.
- Policy on building prerequisites (follow, ask or ignore)? [ask]
Leave this setting set to ask when upgrading CPAN the first time, to prevent dependency lookups automatically installing the latest stable perl. With ask set, one can say no to queries that ask to install Bundle::CPAN, Bundle::libnet, or Data::Dumper until the latest version of CPAN is installed.
- Where is your * program? [*]
This series of questions lets one configure (or disable) paths to various utilities CPAN may use to download or unpack modules. It is safe to leave entries blank, as long as CPAN is left with at least one working program that can download modules and another to expand them.
To disable a particular utility, use a single space instead of a file path.
- Parameters for the 'make install' command?
Set this option to include UNINST=1 if installing as root. This properly cleans out files from older versions of a module being installed. For user-specific setup, I never set the UNINST option, to avoid remove warnings of unremovable system files.
- Now we need to know where your favorite CPAN sites are located.
Choose several CPAN mirror sites. Use traceroute or bandwidth testing tools to determine which mirrors are best for your system. Or, mirror CPAN to a local system.
The mirrors list may need to be updated periodically, if a previous mirror site stops hosting CPAN. Symptoms of this will be slow downloads as CPAN fails though the mirror list.
Reconfiguring CPAN
To alter the CPAN preferences, either edit the Config.pm configuration file manually, or use the following command in the CPAN shell.
cpan> o conf init
The init configuration option runs through all the configuration questions, which may be time consuming. For example, other o conf commands can be used to list, remove, and add mirror sites, and then to save the changes to disk.
cpan> o conf urllist
urllist
ftp://ftp.kernel.org/pub/CPAN/
Type 'o conf' to view configuration edit options
cpan> o conf urllist shift
cpan> o conf urllist push ftp://ftp-mirror.internap.com/pub/CPAN/
cpan> o conf urllist
urllist
ftp://ftp-mirror.internap.com/pub/CPAN/
Type 'o conf' to view configuration edit options
cpan> o conf commit
commit: wrote /usr/local/lib/perl5/5.6.1/CPAN/Config.pm
To manually edit the existing configuration file, either open the user-specific ~/.cpan/CPAN/MyConfig.pm directly, or locate the system-wide configuration file (stored somewhere under the perl @INC path list) to edit with the following command.
$ perl -le 'for (@INC) { $_ .= $ARGV[0]; print if -f }' /CPAN/Config.pm
/System/Library/Perl/CPAN/Config.pm
Usage Tips
Invocation | autobundle | Debugging | Old Modules | Mac OS X | Manual Build | sudo | Uninstalling | Upgrading
Once the latest version of CPAN has been properly installed, it is safe to install Bundle::CPAN to get the latest versions of modules CPAN depends on. If already in the CPAN shell, be sure to run reload cpan first.
cpan> reload cpan
…
cpan> install Bundle::CPAN
The following sections outline various uses and caveats with CPAN.
Invocation
The latest version of CPAN should install a cpan command somewhere on the system, which is easier to use than the traditional means of calling CPAN. With an up to date version of CPAN, the following sets of commands are equivalent. Hereafter, the shorter cpan method will be used.
# enter the CPAN shell
# perl -MCPAN -e shell
# cpan
# install the Acme::Bleach module
# perl -MCPAN -e install Acme::Bleach
# cpan -i Acme::Bleach
Depending on the shell in question, one may need to issue the hash -r or rehash command to make the new cpan command appear in the search path after the latest CPAN is installed. Consult your shell’s documentation for more information on whether this is required.
autobundle
The CPAN autobundle can be used to ease perl upgrades, by creating a special bundle containing all the installed modules of the current version of perl. This bundle can then be installed once the new version of perl is installed.
cpan> autobundle
…
Wrote bundle file
/var/spool/cpan/Bundle/Snapshot_2003_10_01_00.pm
cpan> quit
# upgrade perl here …
# cpan
cpan> install Bundle::Snapshot_2003_10_01_00
Autobundle appears to install the modules in alphabetical order; setting the prerequisites_policy policy to ask should help. Multiple install runs may be needed to install everything properly. Module build failures will lead to much recursion on the part of CPAN.
Debugging
Should tests fail during the make test phase of the module install process, the tests should be investigated to determine their impact. If the failed tests are not relevant to the system in question, the module can be installed with the force option. Otherwise, filing a bug report or otherwise contacting the module author may be in order.
- Force Install
To force install a module, use the force option.
cpan> force install HTML::Mason
- Verbose Testing
More verbose testing will need to be done from the command line. This can be done with the look command from the CPAN shell, along with subsequent shell code as shown in the following example for HTML::Mason, with verbose test logging saved to test.log via the tee(1) command.
cpan> install HTML::Mason
Running install for module HTML::Mason
Running make for D/DR/DROLSKY/HTML-Mason-1.23.tar.gz
…
t/02a-filter..........ok
t/04-misc.............FAILED tests 9-10
Failed 2/11 tests, 81.82% okay
t/05-request..........ok
…
Failed Test Stat Wstat Total Fail Failed List of Failed
---------------------------------------------------------------------------
t/04-misc.t 11 2 18.18% 9-10
2 tests skipped.
Failed 1/23 test scripts, 95.65% okay. 2/376 subtests failed, 99.47% okay.
*** Error code 35
Stop in /var/spool/cpan/build/HTML-Mason-1.23.
/usr/bin/make test -- NOT OK
Running make install
make test had returned bad status, won't install without force
cpan>
cpan> look HTML::Mason
Running look for module HTML::Mason
Trying to open a subshell in the build directory...
Working directory is /var/spool/cpan/build/HTML-Mason-1.23
# make test TEST_VERBOSE=1 TEST_FILE=t/04-misc.t | tee test.log
…
Determination as to whether the failed test is relevant to the system in question will need to be done by a human, typically by running of the test code and consulting the test script and module code to see what is failing. If the failed tests are for features of the module that will not be used, then the module typically can be installed and used without problems.
Old Modules
Some modules found by CPAN are out of date, and will not compile. An updated module may be available on CPAN, just not linked to as “current” by CPAN. Modules known to have this problem include the following. The links below should point to the current version of the modules.
Mac OS X
Problems with CPAN on Mac OS X.
Manual Build
On occasion one may need to manually build or test a module, or consult the module directory to read the documentation on any special needs the module may have. This can be done with the CPAN shell look command.
cpan> look Net::SSLeay
…
Working directory is /var/spool/cpan/build/Net_SSLeay.pm-1.25
# exit
cpan>
sudo
If CPAN is installed in your home directory, the system-wide cpan may cause problems when run via sudo(8), as by default sudo will invoke CPAN using the ~/.cpan/CPAN/MyConfig.pm configuration, but as root, which causes no end of trouble. To avoid this problem, reset the HOME environment variable with the -H option to sudo when invoking CPAN.
$ sudo -H cpan
Uninstalling
CPAN itself offers no uninstall method. Several of the alternatives to CPAN do, such as CPANPLUS or vendor package/port systems. If the module has been fully installed along with a .packlist file for the module, the ExtUtils::Packlist perl module documentation includes code for a modrm command to cleanly remove an installed module.
Modules built with other tools, such as Module::Build, may not write out a .packlist file for use by ExtUtils::Packlist.
Upgrading
How to upgrade all the modules on the system like apt-get is actually documented in the CPAN manual; however, the question comes up often enough that I answer it here as well.
# install everything that is outdated on my disk:
# perl -MCPAN -e 'CPAN::Shell->install(CPAN::Shell->r)'
Problems with this include new versions of modules that may break old module behavior, or various modules that cannot be upgraded due to build failures or system incompatibilities. Best to subject a test system to this sort of mass module upgrade before attempting it on a production system.
if you need to change server pls check this cool site below
CPAN SERVER SITES:
http://www.cpan.org/SITES.html
Wednesday, October 11, 2006
HOW TO SETUP DIAL-IN SERVER on LINUX
Part I . The short version:D on how to get remote access to your Ubuntu box via modem. Install mgetty
sudo apt-get install mgetty
Add a line at the end of file /etc/inittab
S0:2345:respawn:/sbin/mgetty ttyS0
If your modem is on COM1.
Initialize init by typing
sudo init q
Now from Win you can use HyperTerminal to connect to your Ubuntu box with your user name/pass.
For those who want more here is a real deal;)
Part II
The problem: Establish PPP connection to Ubuntu server from a remote location using a modem and to share internet connection.
Laptop modem<> Server modem <> Ethernet Router <> Cable/DSL modem <> Internet
Assumptions: drivers for your modems already installed and modem is at ttyS0. I have an external modem attached to COM1 - ttyS0, COM2 will be ttyS1. I didn't have X installed so I was using nano to edit conf files, you can use gedit if you have X installed or any other editor. Make sure your are familiar with the interface of the editor before you start modifying files. Make sure you create backups of the files before you modify them. You will need to use sudo or login as root for most of the tasks. If you don't know how to configure your dial up on your client look for a dial up howto or Ubuntu docs. I also assume that you already have your local nework including routers properly configured and have an access to the Internet from your Ubuntu server.
Part II.a - Dial in configuration
1. If not already installed use synaptic or apt-get to install ppp and mgetty packages. ppp should be already installed by default so
sudo apt-get install mgetty
2. Create a group ppp by adding a line in file /etc/group ppp:x:1001:
3. Create a new user "pppuser" or whatever you will use for your dial in connection and assign a password by using
sudo adduser pppuser
edit file /etc/passwd or use sudo vipw to change entry for pppuser to pppuser:x:1001:1001:,,,:/home/pppuser:/usr/sbin/ppplogin
4. Add a line to the file /etc/inittab S0:2345:respawn:/sbin/mgetty ttyS0 for modem on ttyS0. Or S1:2345:respawn:/sbin/mgetty ttyS1 for modem on ttyS1 That will let mgetty to accept incoming calls
5.Make a new file /usr/sbin/ppplogin and add the following in there
#!/bin/sh
#/etc/ppp/ppplogin
# PPP login script
mesg n
stty -echo
exec /usr/sbin/pppd -detach modem debug crtscts
6. Set access to the ppplogin file and etc/ppp directory
chmod 750 /usr/sbin/ppplogin
chown root:ppp /usr/sbin/ppplogin
chmod 775 /etc/ppp
chown root:root -R /etc/ppp
7. Restart init by typing init q
If you're use external modem it should be on before that.
8. Open file /etc/mgetty/login.config Comment out everything in there and add a line
/AutoPPP/ - a_ppp /usr/sbin/pppd file /etc/ppp/options
9. Open file /etc/ppp/options and make sure these lines are uncommented. If anything else is uncommented it probably should be commented.
-detach
asyncmap 0
modem
crtscts
proxyarp
lock
require-pap
refuse-chap
ms-dns 192.168.1.1 #put your dns server ip here
usepeerdns
In my case the ms-dns entry had an ip of my router, if you using Linksys router it's 192.168.1.1 by default unless you changed it.
10. Create a file /etc/ppp/options.ttyS0 for the modem on ttyS0 and add following in there
192.168.1.3:192.168.1.201
noauth
Where first address is the address of your server for ppp connection which I think, should be different from your eth ip. The second address is the address that will be assigned to the client when connection is established. It will probably make life easier, unless you know what you doing, if all those addresses on the same subnet as your other computers on the network. (ip starts with the same 192.168.1.x numbers)
You can substitute noauth for a debug line, this way it will log some info about you connection in a syslog.
11. Edit file /etc/ppp/pap-secrets
find a line after # Every regular user can use PPP and has to use passwords from /etc/passwd
It should look something like that
* hostname "" *
substitute hostnatname with * so it looks like that
* * "" *
If you don't do that pap will not authenticate you and you'll be immediately disconnected.
Now you're able to connect using dial-up connection from you laptop or a remote office into your Ubuntu server and use ssh or putty if you're using Win.
Part II.b - Accessing internet from a remote client
I'm sure there other or better solutions to that, but that was easy enough and it worked for me.
sudo apt-get install ipmasq
Done. ipmasq automatically senses all your interfaces and initializes IP Masquerade forwarding/firewalling and allows you to connect to the rest of your network and the Internet.