tag:blogger.com,1999:blog-45194664437401575892012-03-06T00:30:42.869-05:00troy's unix spaceVarious tips and tricks picked up in UNIX engineering.troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.comBlogger1041100tag:blogger.com,1999:blog-4519466443740157589.post-83143002303653343492012-02-26T17:57:00.002-05:002012-02-26T17:59:39.865-05:00vSphere 5 Lab Setup pt 2Continuing from part 1, this post details the configuration on the
management workstation host (glados) and its 2 components, the generic
infrastructure host (lns1) and the management client (fenster).
Our details for this are:
HOSTs: glados, lns1, fenster
OSes: CentOS (6.0 = glados, 6.2 = lns1),
Windows 2008 Server (fenster)
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-37826317922860128162012-02-26T17:06:00.006-05:002012-02-28T12:31:27.079-05:00vSphere 5 Lab Setup pt 1Recently, I decided to setup a vSphere 5 lab at home to help me prepare
for taking the VCP 5 exam. Its purpose is basically to provide me a
setup to get better acquainted with certain vSphere features and maybe
testing of various fault / resolution scenarios. This post is to help me
retain the details of the resulting configuration. Since this is not an
end all / be all lab, if you have any troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-58939429352893604512012-01-30T23:39:00.001-05:002012-01-30T23:50:47.328-05:00Extracting Individual Files from an ArchiveToo often, I only need a single file or a couple of files from an archive.
One answer to this is to just extract the entire archive so that I can
get to those particular files. The more precise answer is to simply
extract only those files that I need and place them where I want them.
The following illustrates a few examples of how to do this with various
archive utlities. Our details for this troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-68853729482702571582012-01-12T15:34:00.001-05:002012-01-12T15:39:44.867-05:00Fixing an Overly Eager chmod in LinuxA while ago, someone asked me how to recover from a mistyped recursive
'chmod' they performed. Similar to the write-up on an "overeager chown",
they mistyped the path and it executed against the root FS (/). Ideally,
one would have a backup to recover from, however that wasn't an option
in either the original situation or the one detailed herein. Our host
details are:
HOSTs: troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-12982312223518103842011-12-28T18:08:00.000-05:002011-12-28T18:08:54.103-05:00updated utility, getldp.pl v0.1.7An updated version of 'getldp.pl', v0.1.7, is now available on the Tools
page. This update adds handling (long overdue) for reaping any child
processes getldp.pl might spawn while listening for CDP / LLDP packets.
Previously, if getldp.pl timed out and exited while waiting on data,
one had to manually kill getldp.pl's children from the process table.
Now it will do the right thing and clean up troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-29209281858508468002011-12-11T17:47:00.000-05:002011-12-11T17:47:09.903-05:00Manual P2V with LinuxWe recently had a situation wherein we were replacing a physical host
with a virtual machine (VM) instance. Rather than simply install a
new VM and reapply software packages and configuration, the need was to
retain the existing installation. To do so, we effectively cloned the
physical host into a VM, which is what the following details. Of note,
while I'm calling this a "manual, offline p2v troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-8037333859887845962011-12-10T16:34:00.000-05:002011-12-10T16:34:27.518-05:00updated utility, getldp.pl v0.1.6An updated version of 'getldp.pl', v0.1.6, is now available on the
Tools page. This version adds details of the sending host including
OS version (CDP only), device capabilities, and management address
(assuming those fields are in the CDP / LLDP packets):
CDP sample:
prompt# getldp.pl -i packet-captures/new.cap -x
Watching for CDP packet on pack-cap/1.pcap for 60 seconds...
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-38798023462257181252011-12-05T15:52:00.000-05:002011-12-05T15:52:43.521-05:00NIC Replacement in LinuxIn the past, I've found it slightly annoying that if I needed to swap
out a network interface card (NIC), Red Hat distros of Linux would kindly
backup the interface configuration for that interface and generate a stub
DHCP configuration in its place. This leaves the host potentially with
an improper network configuration. The following details the necessary
changes on the OS so that the new NICtroyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com2tag:blogger.com,1999:blog-4519466443740157589.post-54847075020317638952011-12-03T22:19:00.000-05:002011-12-03T22:19:42.718-05:00Fixing a Broken initrd in LinuxI recently had a situation where a host's disk controller failed.
The box would get partially through a BIOS post before throwing errors
about the failed controller and then fail to boot. To resolve the
situation, we effectively replaced said controller. This, however, lead
to the situation wherein the initial RAM disk (initrd) didn't load the
appropriate driver / kernel module for the new troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-44319372530416839702011-10-31T00:01:00.004-04:002012-01-12T15:37:25.795-05:00Fixing an Overly Eager chown in LinuxA while ago, someone asked me how to recover from a mistyped recursive
'chown' they performed. They mistyped the path and it executed against
the root FS (/), though they caught it before it acted on everything.
Ideally, one would have a backup to recover from, however that wasn't
an option in either the original situation or the one detailed herein.
Our host details are:
HOSTs:troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com2tag:blogger.com,1999:blog-4519466443740157589.post-41207060376674858382011-09-27T23:40:00.000-04:002011-09-27T23:40:17.979-04:00VxVM DG DisabledDuring a recent SAN maintenance, we ran into an issue on a few hosts that
were unfortunately single pathed to the SAN and lost connectivity to it.
The SAN-presented disk devices were under Veritas Volume Manager (VxVM)
control which started to present I/O errors as a result and displayed
"dgdisabled" in 'vxdisk' output. The following details this situation.
For our setup, we have:
HOST:troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-18615456922508533732011-08-01T02:03:00.001-04:002011-08-01T02:05:49.383-04:00Configuring iSCSI Initiators in Windows 2008Since it's been a while since I've written up a Windows related post, I
thought I'd add one on setting up iSCSI initiators from the command line
(cli). (It also gave me a reason to toy around in PowerShell (PS).)
Microsoft makes iSCSI initiator software available for Windows 2000,
Windows Server 2003, Windows XP, and is included natively for Windows
Server 2008 and Windows Vista. Our host troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com2tag:blogger.com,1999:blog-4519466443740157589.post-26101340663417322022011-07-24T04:07:00.001-04:002011-07-24T04:09:47.211-04:00Process Memory UsageSince resources on a host are finite, it can be useful to know the
impact of a process or processes on those resources. In this instance,
we'll take a look at a process' use of memory. Our details for this are:
HOSTs: cobblepot (Linux)
berkeley (FreeBSD)
sunspot (Solaris)
PROMPT: host [0]
OSes: troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com1tag:blogger.com,1999:blog-4519466443740157589.post-3493243483193214102011-07-13T02:02:00.002-04:002011-09-28T00:01:04.654-04:00Notes on Packages in Solaris 11While working out details for a separate write up, I started stumbling
through Solaris' new handling of packages, IPS. Unlike previous versions
of Solaris which used the traditional "pkgXYZ" commands, directory
structures, etc, IPS (image packaging system) uses a completely new
scheme. The details herein compare some of the surprises I found in
simply trying to identify package ownership of a troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-24988078382367766692011-07-06T03:51:00.000-04:002011-07-06T03:51:20.821-04:00VLAN Tagging and Vanity Naming in Solaris 11While awaiting Solaris 11 to be released, I've recently been poking
around in Solaris 11 Express for a preview of what to expect. In setting
up a test box, I needed to set up VLAN tagging (802.1q trunking) on
an interface. Due to some of the new Solaris 11 networking features,
configuring a VLAN tagged interface is slightly different from what
I recall. Our details this time are:
HOST:troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-25141130879104603002011-06-18T18:11:00.002-04:002011-08-01T02:06:57.494-04:00Configuring iSCSI Initiators in LinuxIn part 1, we worked through how to set up iSCSI targets in Linux
(the server side). Here, in part 2, are my notes on how to set up the
initiators in Linux (the client side). Our host details for this are
the same as part 1:
HOSTS: humboldt (initiator node), cobblepot (target node)
PROMPT: host [0]
OS: CentOS 5.6 Linux (RedHat Clone)
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com2tag:blogger.com,1999:blog-4519466443740157589.post-70414507960778234442011-06-15T23:30:00.001-04:002011-06-18T18:13:42.454-04:00Configuring iSCSI Targets in LinuxSince I've already been looking into iSCSI on Solaris, I thought I'd look
to see what Linux has to offer. While in comparison to Solaris I think
the Linux side could still mature a little, it still has potential and
a functional offering. The following is part 1 of my notes detailing
the configuration of a test Linux iSCSI target node (server side).
Configuration of an iSCSI initiator node (troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-41344218735319681862011-06-13T01:06:00.002-04:002011-08-01T02:07:24.476-04:00Configuring iSCSI Initiators in SolarisIn part 1, we worked through how to set up iSCSI targets in Solaris
(the server side). Now, for part 2, I present my notes on how to set
up the initiators in Solaris on the client side. Our host details for
this are the same as part 1:
HOSTS: sunspot (target node), sunbeam (initiator node)
PROMPT: host [0]
OS: Solaris 10 10/09 (u8)
Before troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-47241896580669364352011-06-08T19:15:00.004-04:002011-06-15T23:32:04.286-04:00Configuring iSCSI Targets in SolarisWhile reviewing potential ideas for several environments, I started
looking into the use iSCSI as a means of providing remote storage.
With that in mind, this is part 1 of my iSCSI notes, detailing the
configuration of a test Solaris iSCSI target node which would provide
storage to client hosts. Configuration of an iSCSI initiator node (client
host) will be detailed part 2. Our host details fortroyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-36891201843916519672011-05-26T23:54:00.002-04:002011-05-27T00:00:52.040-04:00Replicating a ZFS FS Between HostsOccasionally, it can be useful to fully replicate a volume of data from
one host to another. Perhaps you need to mirror a production filesystem
(FS) for use in development, need a sane backup, etc. The following
describes one way of doing this using ZFS. Additionally, we're going
to do so using disparate OSes (though like OSes could have been used).
Here are the setup details:
HOSTS:troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-56769931714521788822011-05-19T23:54:00.003-04:002011-05-20T00:17:46.914-04:00Replacing a Failed SVM DiskAt some point, everyone deal's with a disk failure, as I had to do
recently. In this case, it was a root disk. Thankfully, however,
it was mirrored with SVM (Solaris Volume Manager). Unfortunately, disk
failures aren't the type of thing that should happen too frequently so
it can be easy to overlook steps in the recovery process. The following
details both my oversight and recovery of the troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-34509473461223711852011-05-17T01:12:00.003-04:002011-05-17T01:22:27.799-04:00Sending Messages to UsersRecently while working on a problem with someone else, I needed to have
them run a series of commands with specific options. Since we were
both working remotely, it wasn't as simple as telling them to look
over my shoulder. Though I tried telling them the commands over the
phone, we ran into the issue of parts of the commands being misheard or
misunderstood. To alleviate any confusion, I troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-14445237909271909092011-04-16T17:28:00.003-04:002011-04-16T17:46:18.352-04:00Enabling Local FS Quotas in LinuxQuotas allow you to limit the amount of space a user or group can use
and / or the number of files a user or group can create on a filesystem
(FS). The following shows how to set up these quotas and the expected
user experience. Our host details for this are:
HOST: tux
PROMPT: user@tux [0]
OS: CentOS 5.4 Linux
GROUPS: troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-1360038869640326492011-04-07T22:56:00.002-04:002011-04-07T23:26:44.218-04:00Granting Elevated Privileges in Solarisroot, root, root, everybody always wants root. Developers, application
administrators, users, they all seem to find a reason to "need" root
access. Since normally these needs are for access to particular files
or to perform very specific tasks, only a subset of root's access is
actually needed. File access should be trivial enough, just configure
the appropriate permissions or FACLs. For troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-55317784100228223692011-03-27T23:38:00.002-04:002011-03-27T23:44:33.816-04:00Finding Open Files in LinuxFinding open files in Linux is quite similar to finding them in FreeBSD
and Solaris. Here, we'll specifically identify open files on a particular
filesystem (FS) within a Linux environment. Our host details this
time are:
HOST: cobblepot
PROMPT: cobblepot [0]
OS: CentOS 5.4 Linux
NOTE: The following should also be troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com2tag:blogger.com,1999:blog-4519466443740157589.post-79685703478295297522011-03-25T14:31:00.001-04:002011-03-27T23:39:37.777-04:00Finding Open Files in FreeBSDFollowing up the same in Solaris, here's how to identify open files on
a particular filesystem (FS) in a FreeBSD environment. Our host details
for this are:
HOST: berkeley
PROMPT: berkeley [0]
OS: FreeBSD 6.2
NOTE: The following should likely work on prior FreeBSD
versions and has been verified to troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-17447975979878283792011-03-23T23:29:00.003-04:002011-03-27T23:40:37.215-04:00Finding Open Files in SolarisDuring a recent discussion, another sysadmin asked that I write him a
tool to identify any open files on a particular file system (FS) in a
Solaris environment. Rather than reinvent the wheel, I provided him
with the steps outlined here. Our host details are:
HOST: sunspot
PROMPT: sunspot [0]
OS: Solaris 10 (also works for Solaris 8 and 9)
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-31162688931966669502011-03-17T16:02:00.001-04:002011-03-17T16:17:37.917-04:00updated utility, getldp.plAn updated version of 'getldp.pl', v0.1.5, is now available on the
Tools page. This version now appropriately handles VLAN tagged (802.1q)
CDP packets and includes output for duplex, 802.1q priority, and VLAN ID
(assuming the fields are present in the CDP packets):
snorkle [0] getldp.pl -i packet-captures/new.cap -x
Watching for CDP packet on pack-cap/1.pcap for 60 seconds...
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-5711990502578568052011-03-15T00:34:00.000-04:002011-03-15T00:34:01.942-04:00Converting inet_aton and inet_ntoaAt some point or another, you've probably stumbled upon the 'inet_aton'
and 'inet_ntoa' functions. Perhaps you have used them in programming,
or maybe for a MySQL statement, etc. Briefly, for those not familiar
with either function, 'inet_aton' takes an IP address (10.234.18.76) and
converts it into a packed address (183112268). Alternatively, 'inet_ntoa'
takes a packed address (183112268) andtroyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-9374790274394492132011-03-03T22:53:00.000-05:002011-03-03T22:53:28.077-05:00Configuring NFS in FreeBSDHaving written up Solaris and Linux, it's time to take a look at
configuring NFS in FreeBSD. Briefly, NFS (network file system) provides
access to remote filesystems which appear similar to local resources
on client hosts. The following focuses on simple NFS server and client
configuration in FreeBSD (see note 1). Our host details are:
HOST (server): beastie (10.0.23.181)
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-58288787943222676132011-02-28T02:31:00.001-05:002011-03-03T22:54:33.906-05:00Configuring NFS in LinuxAs a follow up to 'Configuring NFS in Solaris', we turn to the same
in Linux. Briefly, NFS (network file system) provides access to remote
filesystems which appear similar to local resources on client hosts.
The following focuses on simple NFS server and client configuration in
Linux (see note 1). Our host details are:
HOST (server): tux (10.0.23.171)
HOST (client):troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-57268676079834740222011-02-24T16:00:00.006-05:002012-01-12T15:38:29.280-05:00File Integrity Checks via Package DBAfter a system has been installed, files tend to change over time.
These changes may be deliberate, part of normal host operation, the
result of an errant 'chmod', etc. Tools like Samhain, AIDE, and Tripwire
(note 0) have been devised specifically to identify such changes which
they do well. The only problem is that these tools need to be installed
and configured. Though by no means a full troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-70744301077145255482011-02-20T19:43:00.005-05:002011-03-03T22:55:37.803-05:00Configuring NFS in SolarisNetworked environments allow for the sharing of many resources, including
filesystems (FS). As a long time industry standard in UNIX, NFS (network
file system) provides access to remote filesystems which appear similar
to local resources on client hosts. Since Sun Microsystems originally
devised NFS, the following focuses on simple NFS server and client
configuration in Solaris (see note 1). troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-32896213672891144772011-02-17T18:33:00.001-05:002011-02-17T18:35:28.622-05:00Network Ports and Processes (Windows)Previously, I wrote up how to match open network ports to running
processes for FreeBSD, Solaris, and Linux. If UNIX hosts can correlate
ports to processes, shouldn't Windows Server be able to as well?
Beginning with Windows 2000, it can. Our host details for this are:
HOSTS: redmond
PROMPT: redmond C:\>
OS: Windows Server 2008
With Windows troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-61307685898822323352011-02-13T17:29:00.002-05:002011-02-13T17:32:23.269-05:00new utility, basecon.plI've posted a new utility, basecon.pl, to the Tools page. basecon.pl is a simple utility that converts base inputs to selected base output. Currently, basecon.pl can accept input and output types of ascii, binary, decimal, hexadecimal, and octal. The following is sample output from basecon.pl:
user@host [0] basecon.pl -h
basecon.pl converts base inputs to selected base output
Usage: troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-41431570092076365062011-02-03T23:08:00.001-05:002011-02-03T23:10:28.055-05:00Static Routes in LinuxNearly every UNIX sysadmin at some point will need to know how to manage
a host's routing table. With that in mind, here's how to manage static
routes on a Red Hat Linux clone. Our details for this are:
HOST: tux
PROMPT: tux [0]
OS: CentOS 5.4 Linux
In the following exmple, a route is added for network 10.11.18.0/24
(10.11.18.0 - troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-25559441574096585052011-01-31T21:19:00.001-05:002011-02-03T23:11:26.156-05:00Static Routes in FreeBSDA friend of mine recently told me that while he appreciated the detail
and content of my writings, a good deal of the subject matter was a
little beyond him. In support of the fresh-faced sysadmins and those
that might be new to FreeBSD (we were all there once), this write up is
how to add static routes to a FreeBSD host. Our details for this are:
HOST: beastie
PROMPT:troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-78575440568008143862011-01-29T03:48:00.001-05:002011-01-29T03:59:38.453-05:00Breaking and Syncing an MD Root MirrorHere's one of those times where I thought, "I wonder", as in I wonder if
I can break an MD root mirror and sanely recreate it. You can do so with
other software RAID solutions so why not here? Well, you can. There is
nothing short about this procedure, taking 80+ commands to accomplish.
Then again, if you are getting ready to do something to a host that
is potentially going to wreck the root troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-4729273098931725922011-01-23T22:48:00.007-05:002011-01-29T04:02:41.841-05:00Creating an MD Root Mirror in LinuxFor whatever reason, you originally set up your Linux host without a root
disk mirror and now you want it mirrored. One could say "just re-install,"
but surely there is another way, and there is. The solution isn't quick
or simple, though it isn't impossible. Our host details are:
HOST: tux
PROMPTS: [tux[0] |[root@tux /]# |sh-3.2# ]
OS: troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-31362948123350876292011-01-20T01:48:00.000-05:002011-01-20T01:48:05.748-05:00Missing Linux inittabSince I'm somewhat on a "break the boot process" kick, I've deliberately
removed '/etc/inittab' on a Linux host. As the following illustrates,
a missing 'inittab' really isn't as bad as it seems. Our details for
this setup are:
HOST: tux
PROMPTS: [sh-3.2# |tux [0] ]
OS: CentOS 5.4 Linux
We start off with booting the host normally. As troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-76756765298201800712011-01-17T20:31:00.000-05:002011-01-17T20:31:37.331-05:00Missing GRUB Config in LinuxHaving recently written up how to restore GRUB to a missing / corrupted
master boot record (MBR), it seemed appropriate to follow up with
resolving a missing GRUB config file. Our host details for this
situation are:
HOST: tux
PROMPTS: [grub> |tux [0] ]
OS: CentOS 5.4 Linux
DISKS: [sda (hd0|disk1)|sdb (hd1|disk2)]
ROOTtroyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-84267337121507480632011-01-15T17:11:00.001-05:002011-01-15T17:11:43.852-05:00Recreate a Linux EXT3 FSWe've looked at Solaris and FreeBSD, now it's time to look at recreating
a Linux EXT3 filesystem. Similar to the previous two, if you customized
the options to 'mke2fs / mkfs', 'mke2fs -n' is not the answer since it
will only produce data about generic FS creation. Our host details this
time are:
HOST: tux
PROMPT: tux [0]
OS:troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-82315128091510647322011-01-13T19:33:00.001-05:002011-01-15T17:13:22.721-05:00Recreate a FreeBSD UFS FSAs a follow up to recreating a Solaris UFS FS, the following details
FreeBSD. Just as with Solaris, when trying to determine how a filesystem
was created, 'newfs -N /dev/da2s1a' is not the answer, as it only produces
information relevant to a default FS creation. Our host details for this
scenario are:
HOST: beastie
PROMPT: beastie [0]
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-87520240071681505342011-01-11T20:57:00.004-05:002011-01-15T17:12:43.486-05:00Recreate a Solaris UFS FSHave you ever wondered what options were used to create a UFS filesystem
(FS) on a Solaris host? Normally when you ask that question, it
seems someone will refer you to passing '-N' to 'newfs'. Unfortunately,
this response will only provide you with default parameters to build a
generic FS, not the one you are concerned with. Our host details for
this scenario are:
HOST: troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-3153746441588978682011-01-10T02:13:00.004-05:002011-01-29T04:03:30.718-05:00Disk Cloning in LinuxA while back, I wrote up how to clone a disk in Solaris and in FreeBSD,
yet somehow managed to overlook writing up how to do so in Linux.
This write up is to rectify just that, how to clone a disk, in this case,
the root disk, in Linux. The following are our setup details:
HOST: tux
PROMPT: tux [0]
OS: CentOS 5.4 Linux
MASTER DISK:troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-5671236396502369362011-01-08T08:41:00.002-05:002011-01-24T14:06:45.409-05:00GRUB, a Corrupted MBR, and LinuxRecently, after cloning a root disk in Linux, I ran into an issue because
I failed to setup the master boot record (MBR) on the alternate disk.
Everything else was configured, including the boot image files and grub
configuration under /boot/grub, but the MBR wasn't setup. The following
details one solution to this using the following details:
HOST: tux
PROMPTS: troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-76493382251784535412011-01-07T05:56:00.000-05:002011-01-07T05:56:10.784-05:00Adding Swap Space in LinuxWe've discussed FreeBSD and Solaris, now it's time to have a look at
adding swap space in Linux. As previously handled, the setting up
of swap on a free disk partition or else a swap file is discussed.
The following details our example host:
HOST: tux
PROMPT: tux [0]
OS: CentOS 5.4 Linux
PARTITION: If you already have a free partition that troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-60937989188805641292011-01-06T14:54:00.004-05:002011-01-07T05:58:52.232-05:00Adding Swap Space in SolarisFollowing up on "Adding Swap Space in FreeBSD", we take a look at the same
with Solaris. As in the FreeBSD writeup, the usage of a free disk slice
or else a swap file is discussed. The following details our host info:
HOST: snorkle
PROMPT: snorkle [0]
OS: Solaris 10 x86
While detailed is Solaris 10 x86 for our examples, the following should
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-14171315354845052962011-01-05T08:14:00.005-05:002011-01-07T05:58:00.073-05:00Adding Swap Space in FreeBSDWhile it shouldn't occur too frequently, it is likely at some point as a
sysadmin, you will need to add additional swap space to a box. At that
time, your options will be to use a free disk partition with suitable
space or to create a swapfile on an existing filesystem. The following
details both partition and swapfile with the host info below:
HOST: beastie
PROMPT:troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-54043330022460778222011-01-01T18:56:00.006-05:002011-01-05T08:04:58.589-05:00Mounting ISO ImagesSo you've created a fresh ISO image and now you want to use it, verify
its contents, etc. Whatever your purpose, it is fairly trivial to
mount the image file and access it. The following are our host details:
HOSTS: tux (Linux), beastie (FreeBSD), snorkle (Solaris)
PROMPT: host [0]
OSes: CentOS 5.4 (tux)
FreeBSD 8.1 (troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-50118812536472107492010-12-28T05:28:00.002-05:002010-12-28T05:33:09.480-05:00Local Port Forwarding In SSHAnyone familiar with UNIX administration is almost assuredly familiar
with SSH. As a de facto standard of remote access, it can be used as
a replacement to telnet, ftp, rcp, and rlogin. Additionally, it can
be used to provide a secure channel for other needs such encrypting
arbitrary IP services between two hosts or tunnelling IP services
through an intermediary host. The following details howtroyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-44226764698019976192010-12-19T02:45:00.003-05:002010-12-19T02:59:37.998-05:00op, An Alternative To sudo**** What is op? ****
For those unfamiliar with op, one can think of op as being similar
to sudo. As taken from the man page (op.1):
The op tool provides a flexible means for system administrators
to grant trusted users access to certain root operations without
having to give them full superuser privileges. Different sets of
users may access different troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-30046625497280455842010-12-16T19:57:00.005-05:002010-12-19T00:38:04.681-05:00Using Logical Network InterfacesLogical interfaces (logints, vips) are sometimes overlooked though can be
quite handy. For instance, they can be used for IP based hosting (think
webservers), failover of clustered services, testing, providing network
connectivity for zones, etc. Additionally, they are relatively trivial
to setup and use (see note 1). As an aside, in the details below, the
terms logints and vips are used troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-82581880155669311272010-12-12T19:38:00.002-05:002010-12-12T20:06:31.175-05:00BIOS Boot Device - SolarisIn 'Host Info - Boot Parameters', we saw how the boot device on Solaris
could be identified using 'prtconf'. The output was similar to:
boot-device: '/pci@1e,600000/ide@d/disk@0,0:aWhile similar output to the above should exist on any SPARC platform,
on x86 'prtconf' may leave you wanting, returning a result like:
snorkle [0] /usr/sbin/prtconf -vp | /usr/bin/egrep 'boot(args|-troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-29537090578978844192010-12-09T03:52:00.002-05:002010-12-09T04:00:26.950-05:00Mounting an SVM Root Disk from CDROMThe following illustrates how to mount a root disk under Solaris Volume
Manager (SVM) control when booted from a CDROM. Our host details:
HOST: snorkle
PROMPT: cdrom [0]
OS: Solaris 10 u8 x86
SVM ROOT DEVICE: d2
PHYSICAL ROOT SLICE: c1t1d0s0
NOTES: The following is troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-59152621660191230792010-12-07T02:48:00.001-05:002010-12-07T02:52:56.850-05:00FreeBSD Issue with Solaris NFS SharesIn a past NAS migration attempt, it was discovered that FreeBSD boxes
don't necessarily play well with NFS mounts shared from Solaris boxes
when the mount IP is on a virtual interface. The reason behind this
issue is that Solaris, in dealing with virtual interfaces, responds
to incoming requests using one configured IP address on the particular
physical interface. FreeBSD, however, expects a troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-24771869867388047332010-12-03T00:35:00.001-05:002010-12-03T00:38:45.626-05:00Host Info - Windows Server pt. 1What's this, a Windows Server post? He's lost his mind, right?!
This post can be chalked up to "what if X", as in, what if I want to
get host information on a Windows server from the command line. So,
adding to the host info series, here's how to get memory, cpu, and boot
information in Windows 2003. Our host details are:
HOST: fenster
PROMPT: fenster [C:\]
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-34905099551440919902010-11-30T00:25:00.001-05:002010-12-03T00:37:10.962-05:00Host Info - Boot ParametersNow entering the batter's box of the 'host info' series, boot parameters
of a running host. Our lineup of hosts and "stats" are:
HOSTS: snorkle (sunhost), tux (linhost), beastie (bsdhost)
PROMPT: host [0]
Solaris INFO: Solaris 10, x86
Linux INFO: CentOS 5.4, x86
FreeBSD INFO: FreeBSD 8.1, x86
Leading the lineup, and recently tradedtroyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-62521997913269211782010-11-28T17:45:00.000-05:002010-11-28T17:45:47.561-05:00Trust, But VerifyIn setting up this blog, my intention has been to submit posts that are
technically sound and accurate. The posts could then serve me in the
future as reference and hopefully be of service to others. A comment I
received about one of my posts caused me to go back and read thoroughly
every detail of it. Since I had originally written the details almost
2 years ago and only recently posted it troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-67928721880067735272010-11-28T02:19:00.037-05:002010-12-02T20:35:59.146-05:00Replacing VxVM with SVM for the rootdiskThe purpose of this post is to illustrate how to replace Veritas Volume
Manager (VxVM) with Sun Volume Manager (SVM) for control of the root disk
while leaving application volumes under VxVM control. I had originally
tested this against Solaris 9 sparc running VxVM 4.0, though have updated
it for Solaris 10 x86 running VxVM 5.0MP3. The commands are still the
same, with the addition of troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com2tag:blogger.com,1999:blog-4519466443740157589.post-50711091732169093662010-11-24T14:29:00.003-05:002010-11-30T00:25:56.556-05:00Host Info - Network InterfacesAs part of the 'host info' series, we're on to network interfaces.
I know, can't be too hard, run 'ifconfig'. That's fine if you want IP
(layer 3) related information, but what about layer 2 details? To get
link status, associated interface speed, duplex, and VLAN tagging
(802.1q trunking) information, following comes into play. Like before,
the details of our hosts are:
HOSTS:troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com2tag:blogger.com,1999:blog-4519466443740157589.post-80337466402042834852010-11-22T01:54:00.002-05:002010-11-22T02:36:58.579-05:00Overlaid Mount ProblemIn UNIX environments, it's not uncommon to see various file systems (FS)
mounted, be it local storage, fiber attached from a SAN, network based
like NFS, etc. These mounts may be exist for any number of reasons,
such as application data, user home directories, additional log space,
and so on. On occasion, such mounts can obfuscate an otherwise easily
identifiable problem. So here's our host troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-6262251196564286172010-11-19T00:34:00.001-05:002010-11-19T00:37:36.566-05:00VLAN Tagged Interfaces (Linux)A while back, I wrote on how to setup VLAN tagged interfaces on Solaris
and FreeBSD. Though I've also needed to set them up under Linux, I
never seemed to get around to writing up how to do so. To remedy this,
the following are our setup details:
HOST: tux
PROMPT: tux [0]
LINUX DISTRO: CentOS 5.4 (Red Hat EL clone)
INTERFACE: eth0
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-86863741865397749752010-11-17T20:03:00.005-05:002010-11-24T14:34:54.733-05:00Host Info - DisksContinuing the 'host info' series, we check out what disks are presented
to the system. As usual, the details of our hosts are:
HOSTS: snorkle (sunhost), tux (linhost), beastie (bsdhost)
PROMPT: host [0]
Solaris INFO: Solaris 10, x86
Linux INFO: CentOS 5.4, x86
FreeBSD INFO: FreeBSD 8.1, x86
Unlike Linux and FreeBSD, we have to troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-31259728327441366892010-11-15T14:20:00.002-05:002010-11-17T20:04:10.234-05:00Host Info - CPUsFollowing up on memory in the 'host info' series is CPUs. As with memory,
the details of our hosts are:
HOSTS: snorkle (sunhost), tux (linhost), beastie (bsdhost)
PROMPT: host [0]
Solaris INFO: Solaris 10, x86
Linux INFO: CentOS 5.4, x86
FreeBSD INFO: FreeBSD 8.1, x86
CPU information In Solaris:
To determine whether the kerneltroyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-40906902601359277342010-11-13T02:20:00.003-05:002010-11-15T14:25:12.113-05:00Host Info - MemoryThis post on memory is the first in a series on how to retrieve various
bits of host information from Solaris, Linux, and FreeBSD. The OS
details are:
HOSTS: snorkle (sunhost), tux (linhost), beastie (bsdhost)
PROMPT: host [0]
Solaris INFO: Solaris 10, x86
Linux INFO: CentOS 5.4, x86
FreeBSD INFO: FreeBSD 8.1, x86
Of note, most of troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-85980375129823002242010-11-09T22:32:00.002-05:002011-03-27T23:41:48.315-04:00Removing / Recovering an Open File (Linux)Following up on "Removing / Recovering an Open File (Solaris)," the aim
here is to do the same with Linux. The setup is the same, wherein a
'tail' is run in another terminal to hold open a file descriptor to a
file, while said file is accidentally removed. Though the followingdetails CentOS 5.4, the same steps should be applicable on current and
previous versions of like distros.
INFO
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-1107746660314031342010-11-07T03:08:00.004-05:002010-11-07T04:13:12.946-05:00new toolsThe following new utilities have been posted to the 'Tools' page:
getldp.pl -- determine device information for what a network interface is
connected to by way of reading CDP or LLDP packets
ipcalc.pl -- utility for IPv4 range calculations based on CIDRs or netmasks
lprtdiag.pl -- Linux 'prtdiag' like utiltiy
op82dec.pl -- DHCP option 82 suboption 2 decoder / encodertroyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-60839286457415800892010-11-06T03:26:00.002-04:002010-11-17T20:23:10.461-05:00Identifying DMP Subpaths in SolarisRecently I stumbled across one of my notes detailing a quick way to
identify the underlying subpaths supporting VxDMP devices:
frost [0] /usr/sbin/vxdisk path
SUBPATH DANAME DMNAME GROUP STATE
c2t20301E67B00F1032d12s2 c2t20301E67B00F1032d12s2 app6 appdg ENABLED
c3t20311E67B00F1032d12s2 troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-30250528614025262032010-11-03T21:56:00.001-04:002010-11-03T22:09:10.645-04:00Free Space and InodesHave you ever exhausted capacity on a volume when it still shows storage
space available? The following details such a situation where the
intention is to simply copy some man pages into a directory.
HOST INFO
host beastie
prompt beastie [0]
OS FreeBSD 8.1
FILESYSTEM /mnt/somevol
NOTE Though the example troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-49874593168680283922010-11-01T01:57:00.001-04:002010-11-17T20:16:41.961-05:00Logical Disk Names In SolarisAnyone familiar with Solaris instantly recognizes c1t3d0s4 and that
it identifies slice 4 on disk c1t3d0. That's great, but what does the
rest mean. On other UNIX OSes, you might get sdb3 or da2s1a to identify
a disk slice / partition and have to use a variety of means to determine
how it connects to the system. Alternatively, Solaris provides a logical
disk name to easily identify the disk. troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-63856625770880146982010-10-30T02:06:00.005-04:002010-11-22T02:59:43.130-05:00Fix a Broken chmod, or How to chmod chmodSo here's the setup. You've logged into one of your servers to update
permissions on a file only to see this:
server [0] /bin/chmod +x /opt/somefile
-ksh: /bin/chmod: cannot execute [Permission denied]
server [0] /bin/ls -ld /bin/chmod
-rw-r--r-- 1 root root 38564 Jul 18 22:24 /bin/chmod
Ouch, if we use 'chmod' to modify permissions on files, yet 'chmod'
is no troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com4tag:blogger.com,1999:blog-4519466443740157589.post-77080302353725477032010-10-28T15:35:00.000-04:002010-11-17T20:23:10.462-05:00Reassociating a Failed Disk in VeritasRecently we had a situation wherein Vertias Volume Manager (VxVM)
marked a disk as 'failed' while additionally presenting the disk
available for use though unconfigured. The weird part was that VxVM
continued to allow use of the volume until the volume was stopped.
Checks of the disk group (dg) would show 'NODEVICE' for the disk in
question, and a listing of the disk would show no Veritas headertroyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com2tag:blogger.com,1999:blog-4519466443740157589.post-27685926720724179132010-10-21T21:43:00.002-04:002011-03-23T23:34:27.267-04:00Removing / Recovering an Open File (Solaris)While working with filesystems, one will inevitably have to deal with afile that is quickly growing and becoming unmanageable. Unfortunately,this situation normally becomes evident because the space of a volumehas been nearly exhausted. If retention is necessary, you could set upthe file for log rotation and compression. If the data isn't crucial,you could simply zero out the file or remove ittroyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-87692977374162687882010-10-19T04:10:00.001-04:002010-10-30T02:34:18.187-04:00Fun with FACLsQuick, think of a way to give a specific user (jdoe) access to a
particular set of files which are owned by a different user (ivan)
and group (moskau). Depending of what you are trying to accomplish,
I can immediately think of 7 ways:
1: ensure adequate group permissions on the files and assign
jdoe to the necessary group or supplemental group
2: change the group troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-73153599697261664082010-10-19T03:54:00.000-04:002010-11-17T20:23:10.463-05:00Creating a Mirrored Volume with Geom (FreeBSD)INTRO
While it's not uncommon to find a Solaris host with some sort of RAID
configuration in place, within some environments, RAID configurationson FreeBSD seem to almost be held with trepidation. The information
detailed below aims to overcome this hesitancy with the creation of a
simple mirrored (RAID1) volume using FreeBSD's current standard, GEOM.
Though the following details GEOM on troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com2tag:blogger.com,1999:blog-4519466443740157589.post-18041687079899971442010-10-19T03:45:00.010-04:002010-11-17T20:23:10.465-05:00Solaris Devpath EnumerationThe purpose of this doc is show how to physically identify and enumerate
your disk devices on a Sun branded box. The reasons to do this may
include identifying a disk for replacement, determining the bus it is
attached to, I/O balancing, etc. Of note, this doc is only valid with Sun
hardware with Solaris running on it as Sun has been kind enough to release
details regarding such. Sun details troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-89105079761058781072010-10-19T03:03:00.000-04:002010-11-17T20:23:10.466-05:00ZFS Pool and FS CreationVolume and file system management are traditionally separate affairs,
being created and maintained individually of one another. With Sun's
ZFS, the line between the two become blurred. As a brief understanding,
a 'zpool' can be generically considered a volume, comprised of multipledisks and configured in a variety of RAID layouts. A zfs file system is
then created on top of the volume, in the troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-83717183324125892622010-10-19T02:55:00.003-04:002011-03-15T00:47:53.961-04:00Option 82 DecodingWhen looking at DHCP leases, one sometimes may need to be able todecode the option 82 (op82) string passed to the DHCP server. Inthe examples below, the op82 string is provided as the 'relay-agent-remote-id''. The value provided, however, isn'tintuitively clear:
01:01:00:00:ac:08:20:56:22:dd:01:ae
Before going any further, op82dec.pl (on the Tools page) is specifically
capable of troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-15067291590755922122010-10-19T02:22:00.000-04:002010-10-19T02:22:15.812-04:00Rebuilding a Solaris PackageHave you ever needed to rebuild a package already installed on a Solaris
host? Though it may seem to be a rare case, I've had to do this several
times in the past and have yet to find any clear information on it.
A recent instance was where a primary host had a piece of softwareinstalled, though its peer did not. Since the software revision was
older, it was determined to be safer to use the troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com2tag:blogger.com,1999:blog-4519466443740157589.post-69351408251241961342010-10-19T02:07:00.001-04:002011-01-29T04:05:28.059-05:00Breaking and Syncing a SVM Root Mirror (Solaris)INTRO
The information detailed below is to assist with proactive planning for
the potential demise and recovery of a host back to the state it wasin prior to demise. Circumstances this will likely be of benefit would
include the patching of a host, application upgrades by a vendor, trying
new and untested configurations, etc. The value involved is that should
something go catastrophically troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-15324700291528519812010-10-19T01:46:00.001-04:002011-01-29T04:06:42.363-05:00Breaking and Syncing a Hardware Root Mirror (Solaris)INTRO
The information detailed below is to assist with proactive planning for
the potential demise and recovery of a host back to the state it wasin prior to demise. Circumstances this will likely be of benefit would
include the patching of a host, application upgrades by a vendor, trying
new and untested configurations, etc. The value involved is that should
something go catastrophically troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com6tag:blogger.com,1999:blog-4519466443740157589.post-88270729044376430632010-10-19T01:13:00.000-04:002010-10-19T01:13:59.627-04:00Solaris HWADDRs from OBPOccasionally, it's useful to know the hwaddr or MAC addr of an interface
without actually booting a box, such as when you are setting up a
jumpstart config for said box. On Sun Sparc based machines, this
information can be obtained within the OpenBoot (OB) environment.As an aside, until recent years, all network interfaces on a Sun Sparc
box had the same hwaddr, though this was configurable via troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-65249290131652636802010-10-19T01:08:00.004-04:002011-02-28T03:07:09.778-05:00Static Routes in SolarisThis tip may seem somewhat trivial as we all know that to retain static
routes in Solaris, you simply create an init script such as the following:
adler [0] ls -l /etc/rc2.d/S76route -rwxr-xr-x 1 root other 392 Mar 24 2009 /etc/rc2.d/S76route*
adler [0] cat /etc/rc2.d/S76route
#!/bin/sh
/usr/sbin/route add -net 10.150.0.0/16 172.5.21.145
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-19392975421193622612010-10-19T01:00:00.000-04:002010-10-19T01:00:53.386-04:00IPMP and 802.1q on LinuxDescribed herein is the practice of using IPMP (IP MultiPathing) and VLAN
tagged (802.1q trunking) interfaces layered together for the purpose of
port considerations (read: consolidation) and HA (high availability).
The following points should be noted relevant to the host used in thefollowing examples:
OS: Linux (CentOS 5.2)
Kernel: 2.6.18-92.el5
Hosttroyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com2tag:blogger.com,1999:blog-4519466443740157589.post-55020255776398917752010-10-19T00:42:00.002-04:002010-10-19T00:42:56.197-04:00Process EnvironmentsWhen working on a host, it is sometimes of value to know the environmental
settings, such as PATH, that a process is currently using. Fortunately,
there a way of obtaining this information for Solaris, Linux, and FreeBSD.
The following is an example for cron from a Solaris x86 box (though sparccould have been used):
solaris [0] /usr/bin/ps -ef | /usr/bin/grep cron
root 1687troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-26835326081210006792010-10-19T00:38:00.000-04:002010-11-17T20:23:10.470-05:00Recovering from a corrupted ext3 journalOccasionally, an EXT3 volume's journal will become corrupted. When
this occurs, Linux will proactively remount the volume in question as
read-only in attempt to stave off further data corruption, loss, etc.
When the volume is remounted r/o, lines similar to the following can befound in dmesg, written to the console, and possibly in /var/log/messages:
Aug 13 15:29:16 localhost kernel: troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-81905946671927231162010-10-19T00:23:00.004-04:002011-02-24T16:52:27.579-05:00Package Builds in LinuxThis document details how to create a package on Linux. Only the creationof a simple package is detailed in this doc. For more elaborate packages,see the relevant man pages.The following assumptions are being made regarding the package builddescribed herein: - the program, etc, has already been installed on the host the package is to be built upon - program troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-80093307106296293722010-10-19T00:15:00.011-04:002010-10-19T00:34:46.552-04:00Package Builds in FreeBSDThis document details how to create a package on FreeBSD. Only thecreation of a simple package is detailed in this doc. For more elaboratepackages, see the relevant man pages.The following assumptions are being made regarding the package builddescribed herein: - the program, etc, has already been installed on the host the package is to be built upon - program troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-33619450508216902262010-10-18T23:04:00.002-04:002010-10-18T23:34:54.188-04:00Package Builds in SolarisThis document details how to create a package on Solaris. Only the
creation of a simple package is detailed in this doc. For more elaborate
packages, see the relevant man pages.
The following assumptions are being made regarding the package build
described herein:
- the program, etc, has already been installed on the host
the package is to be built upon
- for troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-78808227462891444992010-10-18T22:52:00.001-04:002011-02-17T18:34:42.795-05:00Network Ports and ProcessesLooking at a host, there are times when it may be indeterminablewhat process may be holding a particular port open. Perhaps becausethere is no entry for the port in /etc/services, thus netstat, etc,will not display a recognizable name. The reason may be that the portnumber is not identifiable relative to any process running on the host.For whatever the reason, it may be necessary to determine troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-51538582740684121312010-10-18T22:40:00.000-04:002010-10-18T22:40:41.144-04:00Strange FilenamesInevitably, while working on a server, one will come across a file with a
strange name that seemingly cannot be easily operated upon. Some examples
of commonplace files tend to be "!" and "@", wherein the "strange" file
was created while working perhaps in vi. Other examples may be "--", "*"," . ^ ", " ", "$", "|", etc. These file present somewhat of a
challenge as they have the troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-77293290843967047252010-10-18T22:35:00.000-04:002010-10-18T22:35:03.073-04:00Simple Pattern ReplacementsManaging data within a file sometimes requires that all instances of a
specific word / phrase be replaced with that of another. This could be
something along the lines of a file from one host is copied to another,
thus the hostname contained in the file must be updated to reflect the
new host it resides upon. Another instance could be the removal of ^M
from within a file originally dos troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-19398046469322194002010-10-18T22:29:00.002-04:002010-10-19T01:09:13.086-04:00Solaris Management FacilityWith the advent of Solaris 10, Sun introduced a new mechanism to manage
the running of services, the services management facility (smf(5)).
This is important to note as it impacts the startup of services
traditionally handled via the various rcX.d directories. In short, whilethe rcX.d directories are still existent and may be used for "legacy"
scripts, the actual handling of the majority of the troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-66781722248703797302010-10-18T22:18:00.002-04:002011-07-13T02:03:42.042-04:00Package File OwnershipOnce a system is built and begins to be used, undoubtedly, more files
are added to the system. These files may be logs, user / application
created files, files that are part of a package, or those compiled
from source. When poking around a system, one may stumble across a filethat was installed by a package though either the package is unknown to
the individual or is not readily identifiable. troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-3569162758586037882010-10-18T22:11:00.000-04:002010-10-18T22:11:38.613-04:00File TimesOccasionally it may be necessary to determine the various times of a file.These times break down to access (when the file was last accessed),modification (time of last modification), and inode (time of lastmodification of the inode (mode / permission changes, etc.). FreeBSD,Solaris, and Linux have various means of determining this information,one of which being /bin/ls. The output from this troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-22560687931916581122010-10-18T21:55:00.000-04:002010-10-18T21:55:34.044-04:00Epoch TimeHave you ever needed a way to determine UNIX (Epoch) Time, or perhaps its
counterpart in 24 hour format?
via perl:
server [0] perl -e 'print scalar time;' ; echo
1154020678
To use UNIX time to determine it's correlative in 24 hour format:
server [0] perl -e 'print scalar localtime 1154020678 ;' ; echo
Thu Jul 27 13:17:58 2006
via gnu awk:
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com2tag:blogger.com,1999:blog-4519466443740157589.post-1393901491534539412010-10-18T21:51:00.003-04:002011-01-10T02:25:05.925-05:00Disk Cloning in FreeBSDDisk Cloning in FreeBSD:
A few assumptions:
master disk: da0
\--> disk already configured and in use
alternate: da1
\--> disk not configured and will become a clone of master
configured slice: s1
\--> slices 2, 3, and 4 remain unconfigured
configured partitions: a, b, d, e
\--> a=/, b=swap, d=/troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-77994051230021935972010-10-18T21:34:00.003-04:002011-03-15T00:48:06.287-04:00Base ConversionsEvery now and then, it may be necessary to do base conversions on numbers.
While some may be readily able to read 0x17 in hexadecimal as 23 in
decimal as 10111 in binary as 27 in octal, there are those who may not.
A simple example of how this information may be of use would be in
figuring out an interface's netmask from ifconfig's output. In this
output, the netmask is printed in hexadecimal:
troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0tag:blogger.com,1999:blog-4519466443740157589.post-66066927023690075922010-10-18T21:27:00.001-04:002010-11-17T20:23:53.707-05:00Determining Partition TablesThis tip deals somewhat with Solaris, though moreso with FreeBSD.
On a Solaris host, to quickly determine how a hard disk may be carved up,
one can simply run prtvtoc against the device: sunbox [0] /usr/bin/uname -a
SunOS sunbox 5.10 Generic_118855-14 i86pc i386 i86pc
sunbox [0] /usr/sbin/prtvtoc /dev/dsk/c1t0d0s2
* /dev/dsk/c1t0d0s2 partition map
*
* Dimensions:
* troyhttp://www.blogger.com/profile/02867962932926227365noreply@blogger.com0