Re: Quick DNS perfromance measurement trick
I tried the specified command from three different sites and they all gave essentially identical responses: "aaa.bbb.ccc.ddd is POOR: 26 queries in 3.1 seconds from 1 ports with std dev 0.00" That aaa.bbb.ccc.ddd address seems to be the (possibly NAT'd) IP addr that the target site sees mentioned in the inbound packets; I have no idea about the rest of it... ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: How to troubleshoot wide area network performance problem?
Alex Hewitt wrote: > I have clients with an interesting network problem. One location in > Bedford New Hampshire using a fractionated T1 has routinely been > transmitting studies to an office in Nashua New Hampshire. There have > been no problems with this for at least 18 months. However recently > (about a week ago), the transmissions suddenly became slow, really slow. > A transmission that was taking around 10 minutes suddenly jumped to 2-3 > hours. The customer in Bedford New Hampshire is using One > Communications. So far I haven't asked them to look at this problem > because I've been trying to clarify it. The office in Nashua has > Comcast business class service with a static IP address. > > Here's where it gets interesting. I had the Bedford client transmit the > data to my system in Manchester New Hampshire. I have Comcast > residential service. The data usually takes about 8 minutes to arrive at > my location. I then send the data to the Nashua office and it typically > takes 25-30 minutes. The payload is a collection of images that are > typically between 65 and 70 MB. > That sounds like typical asymmetric cable modem connection. > Today Comcast at the request of the customer sent someone on site to the > Nashua site. The tech did some speed tests using the DSLReports > Speakeasy test suite. He was getting 20 mbs down, 3+ mbs up which is > pretty decent. For the fun of it I had him download a 47 MB antivirus > program. His first try was ridiculous telling him it was going to take 4 > + hours. I had him break the connection and try again and this time the > download took around a minute. > Its hard to tell if that problem was on the server end or some router between the local and remote system. > And it gets more interesting...another client in Salem New Hampshire > needed to send their data to the Nashua site (they use Verizon DSL). It > arrived in about 8 minutes. > This would imply the Nashua site is OK. > So my Comcast connection which is fairly decent is taking a half hour to > send 65-70 MB to the Nashua site. The Salem site is taking 8 minutes for > something approximately the same size and the Bedford site is taking > several hours. > Before paying for a tech to go to the Bedford site, I would try a *short* flood ping to the ISP's first advertised router (short = 5 seconds) and see what sort of loss you get. This will tell you if the problem is in the ISP's on-site equipment (and if so, the tech can diagnose it). Then try pinging to the first router outside of the ISP's network. This should tell you if the problem is inside/outside the ISP network. Armed with this info, you can then call the Bedford ISP and ask them what's going on. > Traceroute doesn't show much interesting - it craps out after the first > 5 hops. Pinging (standard payload) from my office to the Nashua site is > averaging less than 20 ms. One odd thing is that when I'm in the process > of sending data to the Nashua site my pings jump up to 650 - 800 ms. > > The Comcast tech was happy to conclude that the Nashua site was working > properly. They checked transmission levels, noise and of course the guy > downloaded some files and ran the Speakeasy speed tests and all of that > looked good. > > Any ideas how to proceed on a problem like this? Currently I'm having > the customer transmit their data to me and then I re-transmit because my > connection although slow is probably 4 or 5 times faster than theirs. > Sounds to me like the Bedford ISP/Carrier needs a clue bat. --Bruce ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: How to troubleshoot wide area network performance problem?
On Thu, 2008-07-10 at 19:36 -0400, Alex Hewitt wrote: > I have clients with an interesting network problem. One location in > Bedford New Hampshire using a fractionated T1 has routinely been > transmitting studies to an office in Nashua New Hampshire. There have > been no problems with this for at least 18 months. However recently > (about a week ago), the transmissions suddenly became slow, really slow. > A transmission that was taking around 10 minutes suddenly jumped to 2-3 > hours. The customer in Bedford New Hampshire is using One > Communications. So far I haven't asked them to look at this problem > because I've been trying to clarify it. The office in Nashua has > Comcast business class service with a static IP address. > > Here's where it gets interesting. I had the Bedford client transmit the > data to my system in Manchester New Hampshire. I have Comcast > residential service. The data usually takes about 8 minutes to arrive at > my location. I then send the data to the Nashua office and it typically > takes 25-30 minutes. The payload is a collection of images that are > typically between 65 and 70 MB. > > Today Comcast at the request of the customer sent someone on site to the > Nashua site. The tech did some speed tests using the DSLReports > Speakeasy test suite. He was getting 20 mbs down, 3+ mbs up which is > pretty decent. For the fun of it I had him download a 47 MB antivirus > program. His first try was ridiculous telling him it was going to take 4 > + hours. I had him break the connection and try again and this time the > download took around a minute. > > And it gets more interesting...another client in Salem New Hampshire > needed to send their data to the Nashua site (they use Verizon DSL). It > arrived in about 8 minutes. > > So my Comcast connection which is fairly decent is taking a half hour to > send 65-70 MB to the Nashua site. The Salem site is taking 8 minutes for > something approximately the same size and the Bedford site is taking > several hours. > > Traceroute doesn't show much interesting - it craps out after the first > 5 hops. Pinging (standard payload) from my office to the Nashua site is > averaging less than 20 ms. One odd thing is that when I'm in the process > of sending data to the Nashua site my pings jump up to 650 - 800 ms. > > The Comcast tech was happy to conclude that the Nashua site was working > properly. They checked transmission levels, noise and of course the guy > downloaded some files and ran the Speakeasy speed tests and all of that > looked good. > > Any ideas how to proceed on a problem like this? Currently I'm having > the customer transmit their data to me and then I re-transmit because my > connection although slow is probably 4 or 5 times faster than theirs. > > -Alex > > A few more bits of information - I replaced the router in the Nashua office (Netgear FVS 114) with a new identically configured model. The download performance and speed tests were run with the Netgear router in place (all good). I disconnected the router from the cable modem and hooked the Mac that runs the client application directly to the cable modem. Again all download tests look normal. I replaced the original Mac with a newer model. The old system was a Mac Mini with 1 GB of Ram and a G4 CPU. The replacement model was a dual core Intel based Mini with 2 GB of Ram. The new system is definitely snappier but doesn't affect the problem at all. -Alex > > > ___ > gnhlug-discuss mailing list > gnhlug-discuss@mail.gnhlug.org > http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/ ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
How to troubleshoot wide area network performance problem?
I have clients with an interesting network problem. One location in Bedford New Hampshire using a fractionated T1 has routinely been transmitting studies to an office in Nashua New Hampshire. There have been no problems with this for at least 18 months. However recently (about a week ago), the transmissions suddenly became slow, really slow. A transmission that was taking around 10 minutes suddenly jumped to 2-3 hours. The customer in Bedford New Hampshire is using One Communications. So far I haven't asked them to look at this problem because I've been trying to clarify it. The office in Nashua has Comcast business class service with a static IP address. Here's where it gets interesting. I had the Bedford client transmit the data to my system in Manchester New Hampshire. I have Comcast residential service. The data usually takes about 8 minutes to arrive at my location. I then send the data to the Nashua office and it typically takes 25-30 minutes. The payload is a collection of images that are typically between 65 and 70 MB. Today Comcast at the request of the customer sent someone on site to the Nashua site. The tech did some speed tests using the DSLReports Speakeasy test suite. He was getting 20 mbs down, 3+ mbs up which is pretty decent. For the fun of it I had him download a 47 MB antivirus program. His first try was ridiculous telling him it was going to take 4 + hours. I had him break the connection and try again and this time the download took around a minute. And it gets more interesting...another client in Salem New Hampshire needed to send their data to the Nashua site (they use Verizon DSL). It arrived in about 8 minutes. So my Comcast connection which is fairly decent is taking a half hour to send 65-70 MB to the Nashua site. The Salem site is taking 8 minutes for something approximately the same size and the Bedford site is taking several hours. Traceroute doesn't show much interesting - it craps out after the first 5 hops. Pinging (standard payload) from my office to the Nashua site is averaging less than 20 ms. One odd thing is that when I'm in the process of sending data to the Nashua site my pings jump up to 650 - 800 ms. The Comcast tech was happy to conclude that the Nashua site was working properly. They checked transmission levels, noise and of course the guy downloaded some files and ran the Speakeasy speed tests and all of that looked good. Any ideas how to proceed on a problem like this? Currently I'm having the customer transmit their data to me and then I re-transmit because my connection although slow is probably 4 or 5 times faster than theirs. -Alex ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
RE: Example of ARM based linux board using initramfs and serial console
On Thu, July 10, 2008 4:08 pm, Alexander Wolfson said: > This is our own board based on our own chip which among other things has > ARM926EJ-S (ARMv5TEJ) core. > There is no BSP yet, no Flash or USB drivers - only limited access to > the board over JTAG. We can boot the board up to the point when kernel > dies because there is no init. This why I need initramfs now. > While i don't know the details, I understand that it's common practice when porting to a new platform to use bash, or perhaps a simpler shell, in place of init, as a first step to achieve an initial boot. -- John Abreau / Executive Director, Boston Linux & Unix IM: [EMAIL PROTECTED] / [EMAIL PROTECTED] / [EMAIL PROTECTED] / [EMAIL PROTECTED] Email [EMAIL PROTECTED] / WWW http://www.abreau.net / PGP-Key-ID 0xD5C7B5D9 PGP-Key-Fingerprint 72 FB 39 4F 3C 3B D6 5B E0 C8 5A 6E F1 2C BE 99 -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
RE: Example of ARM based linux board using initramfs and serial console
Thank you, I never heard of that one before Alex -Original Message- From: Thomas Charron [mailto:[EMAIL PROTECTED] Sent: Thursday, July 10, 2008 5:44 PM To: Alexander Wolfson Cc: [EMAIL PROTECTED]; gnhlug-discuss@mail.gnhlug.org Subject: Re: Example of ARM based linux board using initramfs and serial console On Thu, Jul 10, 2008 at 5:41 PM, Thomas Charron <[EMAIL PROTECTED]> wrote: > The tool: Linux Target Image Builder http://www.bitshrine.org/ I should elaborate. the Linux Target Image Builder is actually not done by Freescale, it's an open source utility whos website is located at http://savannah.nongnu.org/projects/ltib -- -- Thomas ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Example of ARM based linux board using initramfs and serial console
On Thu, Jul 10, 2008 at 5:41 PM, Thomas Charron <[EMAIL PROTECTED]> wrote: > The tool: Linux Target Image Builder http://www.bitshrine.org/ I should elaborate. the Linux Target Image Builder is actually not done by Freescale, it's an open source utility whos website is located at http://savannah.nongnu.org/projects/ltib -- -- Thomas ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Example of ARM based linux board using initramfs and serial console
On Thu, Jul 10, 2008 at 5:13 PM, Alexander Wolfson <[EMAIL PROTECTED]> wrote: > Thanks, > I looked at the docs - unfortunately they don't have details I am > looking for - it is more a cook book. Source files + the davinci pdf > would be great. > I have seen a Gentoo's already. It uses a complicated build system so > details a sort of hidden. If you're looking for a BSP to actually look at, Freescale has several available, as well as a tool they utilize to allow customers using freescale chips (including ARM) to build their distros. The BSPs: http://www.freescale.com/webapp/sps/site/overview.jsp?nodeId=02VS0l320822D0033202A7 The tool: Linux Target Image Builder http://www.bitshrine.org/ -- -- Thomas ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
RE: Example of ARM based linux board using initramfs and serial console
Thanks, I looked at the docs - unfortunately they don't have details I am looking for - it is more a cook book. Source files + the davinci pdf would be great. I have seen a Gentoo's already. It uses a complicated build system so details a sort of hidden. Alex -Original Message- From: Thomas Charron [mailto:[EMAIL PROTECTED] Sent: Thursday, July 10, 2008 4:54 PM To: Alexander Wolfson Cc: [EMAIL PROTECTED]; gnhlug-discuss@mail.gnhlug.org Subject: Re: Example of ARM based linux board using initramfs and serial console On Thu, Jul 10, 2008 at 4:41 PM, Alexander Wolfson <[EMAIL PROTECTED]> wrote: > Is it possible to take a look at the code for DaVinci? > At least kernel configuration file and initramfs directory structure. > Or you have to pay 7K for the license? > I am afraid that I am missing some nuance, so I would like to have some > sanity check before starting a full blown debugging. Take a look at http://focus.ti.com/lit/an/spraah2a/spraah2a.pdf I think it has basically the 'Idiots Guide' that they use as an example. Another thing to read that may help is http://gentoo-wiki.com/HOWTO_Custom_Initramfs_From_Scratch -- -- Thomas ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Quick DNS perfromance measurement trick
Just saw this on a DNS forum, It seems to work nicely - I hope every one tries it and reports their result here in the gnhlug list :-) found here : http://lists.oarci.net/pipermail/dns-operations/2008-July/002932.html Here is the command: dig +short porttest.dns-oarc.net TXT Here are my results: z.y.x.w.v.u.t.s.r.q.p.o.n.m.l.k.j.i.h.g.f.e.d.c.b.a.pt.dns-oarc.net. "209.244.7.43 is POOR: 38 queries in 1.9 seconds from 2 ports with std dev 0.94" It appears that "good" resolvers have lots of ports. Anyone who wants to take a whack at explaining what this means is very welcome! That IP above is not known to me - here is my /etc/resolv.conf: ; generated by /sbin/dhclient-script search hsd1.ma.comcast.net. nameserver 68.87.71.226 nameserver 68.87.73.242 Jeff Kinz ( ... OR his evil twin ) -- "When fascism comes to America it will be wrapped in a flag and carrying a cross." - Sinclair Lewis. Don't you hate it when a prognosticator is right? ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Example of ARM based linux board using initramfs and serial console
On Thu, Jul 10, 2008 at 4:41 PM, Alexander Wolfson <[EMAIL PROTECTED]> wrote: > Is it possible to take a look at the code for DaVinci? > At least kernel configuration file and initramfs directory structure. > Or you have to pay 7K for the license? > I am afraid that I am missing some nuance, so I would like to have some > sanity check before starting a full blown debugging. Take a look at http://focus.ti.com/lit/an/spraah2a/spraah2a.pdf I think it has basically the 'Idiots Guide' that they use as an example. Another thing to read that may help is http://gentoo-wiki.com/HOWTO_Custom_Initramfs_From_Scratch -- -- Thomas ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
RE: Example of ARM based linux board using initramfs and serial console
Is it possible to take a look at the code for DaVinci? At least kernel configuration file and initramfs directory structure. Or you have to pay 7K for the license? I am afraid that I am missing some nuance, so I would like to have some sanity check before starting a full blown debugging. -Original Message- From: Michael Nolin [mailto:[EMAIL PROTECTED] Sent: Thursday, July 10, 2008 4:32 PM To: gnhlug-discuss@mail.gnhlug.org; Alexander Wolfson Subject: Re: Example of ARM based linux board using initramfs and serial console > > I would like to find a working example of initramfs on the > ARM based > board. > > > > > > > I am porting Linux to the ARM 926 based board. There are no > network or > Flash drivers available yet. No LCD is available yet as > well > > I tried to use initramfs just to test the board , etc. > > This is a new chip so hardware bugs are possible. > The Texas Instruments Davinci platform uses a ARM926 core. U-Boot with a Monta Vista Pro Linux kernel. I would start with U-Boot/arch/ARM to get started on a new evaluation platform. This platform was not free about 7K for the licenses, I thought this was a good value considering rolling your own from uCLinux.org is work. The hardware was only $500.00 I'm using an MMU less ARM966 with no OS, GNU Linux development environment . ARM embedded question might best be answered on some of the comp.arch.embedded.. GNUARM Yahoo groups? Michael Nolin Embedded Solutions Unlimited, LLC ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Example of ARM based linux board using initramfs and serial console
> > I would like to find a working example of initramfs on the > ARM based > board. > > > > > > > I am porting Linux to the ARM 926 based board. There are no > network or > Flash drivers available yet. No LCD is available yet as > well > > I tried to use initramfs just to test the board , etc. > > This is a new chip so hardware bugs are possible. > The Texas Instruments Davinci platform uses a ARM926 core. U-Boot with a Monta Vista Pro Linux kernel. I would start with U-Boot/arch/ARM to get started on a new evaluation platform. This platform was not free about 7K for the licenses, I thought this was a good value considering rolling your own from uCLinux.org is work. The hardware was only $500.00 I'm using an MMU less ARM966 with no OS, GNU Linux development environment . ARM embedded question might best be answered on some of the comp.arch.embedded.. GNUARM Yahoo groups? Michael Nolin Embedded Solutions Unlimited, LLC ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
RE: Example of ARM based linux board using initramfs and serial console
This is our own board based on our own chip which among other things has ARM926EJ-S (ARMv5TEJ) core. There is no BSP yet, no Flash or USB drivers - only limited access to the board over JTAG. We can boot the board up to the point when kernel dies because there is no init. This why I need initramfs now. Alex -Original Message- From: Thomas Charron [mailto:[EMAIL PROTECTED] Sent: Thursday, July 10, 2008 3:59 PM To: Alexander Wolfson Cc: gnhlug-discuss@mail.gnhlug.org Subject: Re: Example of ARM based linux board using initramfs and serial console On Thu, Jul 10, 2008 at 2:59 PM, Alexander Wolfson <[EMAIL PROTECTED]> wrote: > Hi all, > I would like to find a working example of initramfs on the ARM based board. > I am porting Linux to the ARM 926 based board. There are no network or Flash > drivers available yet. No LCD is available yet as well I'm confused. You're trying to make a full Linux system on an Arm-926 based board which doesn't already have a BSP which includes Linux? The kernel itself has run on the 926 for years. -- -- Thomas ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Example of ARM based linux board using initramfs and serial console
On Thu, Jul 10, 2008 at 2:59 PM, Alexander Wolfson <[EMAIL PROTECTED]> wrote: > Hi all, > I would like to find a working example of initramfs on the ARM based board. > I am porting Linux to the ARM 926 based board. There are no network or Flash > drivers available yet. No LCD is available yet as well I'm confused. You're trying to make a full Linux system on an Arm-926 based board which doesn't already have a BSP which includes Linux? The kernel itself has run on the 926 for years. -- -- Thomas ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
NoScript Error
I have the NoScript extension installed on SeaMonkey. I just upgraded from 1.1.9 to 1.1.19 and had to reinstall all my addons. I am now getting errors even on blank pages. [NoScript] nsBrowserAccess not found?! What does this mean? How can I fix it? Any thoughts? More info: I open seamonkey and it's always set to open to a blank page. After upgrading and reinstalling NoScript and Adblocker Plus I close the browser and re-open it. Now I have a error panel open at the bottom of the screen. Checking for error messages I find 2: Error: [Exception... "Component returned failure code: 0x8000 (NS_ERROR_UNEXPECTED) [nsIPrefBranch.getCharPref]" nsresult: "0x8000 (NS_ERROR_UNEXPECTED)" location: "JS frame :: chrome://imagezoom/content/ZoomImageManager.js :: ZoomImageManager :: line 5" data: no] Source File: chrome://imagezoom/content/ZoomImageManager.js Line: 5 and Error: [Exception... "Component returned failure code: 0x8000 (NS_ERROR_UNEXPECTED) [nsIPrefBranch.getCharPref]" nsresult: "0x8000 (NS_ERROR_UNEXPECTED)" location: "JS frame :: chrome://imagezoom/content/overlay.js :: initImageZoom :: line 45" data: no] Source File: chrome://imagezoom/content/overlay.js Line: 45 When I click on the Source File link from the first error I get the first message of: [NoScript] nsBrowserAccess not found?! It also pops up a new Error message: Error: [Exception... "Component returned failure code: 0x80004005 (NS_ERROR_FAILURE) [nsIWebProgress.removeProgressListener]" nsresult: "0x80004005 (NS_ERROR_FAILURE)" location: "JS frame :: chrome://global/content/bindings/browser.xml :: removeProgressListener :: line 325" data: no] Source File: chrome://global/content/bindings/browser.xml Line: 325 Clicking the Source File link from the second error give me the second message of: [NoScript] nsBrowserAccess not found?! And it pops up *another* error message: Error: [Exception... "Component returned failure code: 0x80004005 (NS_ERROR_FAILURE) [nsIWebProgress.removeProgressListener]" nsresult: "0x80004005 (NS_ERROR_FAILURE)" location: "JS frame :: chrome://global/content/bindings/browser.xml :: removeProgressListener :: line 325" data: no] Source File: chrome://global/content/bindings/browser.xml Line: 325 I'm running seamonkey-1.1.10-1.fc8 on Fedora release 8 (Werewolf) So, any thoughts? Tom -- TARogue (Linux user number 234357) -Give a man a fish & he's fed for a day. Teach him to fish & he'll spend all day drinking beer getting sunburned. ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Example of ARM based linux board using initramfs and serial console
Hi all, I just subscribed to the list and would like to say hello to everybody. I would like to find a working example of initramfs on the ARM based board. I am porting Linux to the ARM 926 based board. There are no network or Flash drivers available yet. No LCD is available yet as well I tried to use initramfs just to test the board , etc. This is a new chip so hardware bugs are possible. Unfortunately I never used intramfs before and was looking for a good doc or better an example. The closes I found beyond "/Documentation/filesystems/ ramfs-rootfs-initramfs.txt was on free-electrons.com . They were using framebuffer and Qemu no serial console I built initramfs file system using "buildroot" project. So far board hangs after getting to the point where I assume it tries to go to initramfs. I started instrument the kernel, etc. to see what is going on, but it would be nice to have a working example to know that at least I did not make any stupid mistake. Thanks, Alex ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Adding a new drive / fstab
On Thu, Jul 10, 2008 at 11:33 AM, V. Alex Brennen <[EMAIL PROTECTED]> wrote: > The reason that many people avoid putting anything under '/' that is not > created by the operating system itself is that if you have any type of > problem mounting the disk space that you plan to use under that > directory, your programs or system can fill the '/' partition ... The same would be true of /mnt/, of course, unless that is its own filesystem separate from the root. And on many (most?) modern Linux distros, the default is for one big partition for everything. If one sets permissions on the underlying mount point such that write access is denied, you could solve that problem without resorting to a separate partition just for mount points. "chattr +i /mnt", for example. Myself, I haven't worried about it much, but I haven't had to deal much with different filesystem storages on the same host. If one filesystem isn't mounting, it will generally be noticed immediately, because all sorts of things won't be working. If I had big storage appliances or lots of cross-mounts, like you described, my story would likely be different. "Experience is what you get by not having it when you need it" and all that. :) I *do* tend to create separate filesystems, but for other reasons. I'm most concerned with a runaway process filling up the filesystem it is *supposed* to be writing to. For example, a malfunctioning process generating tons of log messages can fill up the log partition. (Which can in turn cause all sorts of processes to hang/crash.) So I isolate them for that reason. It also localizes filesystem fragmentation, and allows for more selective use of the "noexec" option to "mount" as a defense against malware. I use LVM rather than simple partitions to make resizing/reallocation easier. For example, on the GNHLUG server, there are LVs for / (root), swap, /usr, /usr/local, /home, /sites, /tmp, /var/, /var/log, and /var/spool. > Linux has become somewhat resistant to these types of problems. I can say from recent experience that a full filesystem can impair a modern Linux system pretty effectively. :-/ Some of the fancier filesystems don't use inodes; I expect they'd be immune from inode exhaustion. But they can still run out of space, of course. > Many default disk partitioning plans, including those of RedHat and many > others, create a very small '/' partition. Red Hat (and derivatives) hasn't done things that way for years. Their default scheme is to create a small /boot/ partition, and then create one big / partition for everything else. The reason for that is *that* various bootloaders, motherboards, BIOSes, etc., have had issues booting from large disks. Ubuntu is the same way, as I recall. -- Ben ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Please trim quoted text (was: General Procedure)
> Not to nitpick, but we will still have to trim quotes. > > Quote collapsing is great for the readers, but it won't help with message > size limits on the server :-) > I'm not suggesting that people don't trim their quotes, just that there are clientside approaches to ameliorating the visual, if not architectural, issues. Take it easy, David Berube ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Adding a new drive / fstab
Many people said to just "use what works for you." I agree. There's no deep need to conform to the POSIX HFS standard. Especially, since many of the points in the standard come from negotiation with the makers of flavors of UNIX which are now long dead or dying. If you look at the LSB equivalent of the HFS standard, you can see it is much less verbose and allows much more freedom for the administrator. Regarding creating an additional mount point in the root directory, which was also suggested, there are reasons why you may not want to do that. The reason that many people avoid putting anything under '/' that is not created by the operating system itself is that if you have any type of problem mounting the disk space that you plan to use under that directory, your programs or system can fill the '/' partition (either its available blocks or its available inodes). On some systems, this can cause booting and operational problems with your server bringing some or all of your service down. Linux has become somewhat resistant to these types of problems. It is much more so than the older *NIX Operating Systems. But, on Linux you would still encounter some problems. Depending on how your OS Disk(s) are partitioned, those problems could be very serious. This issue with filling up disk partitions for which parts of your core operating system are depending on being able to write to has been written about extensively. The majority of the content on the web comes from security sites and deals with including /var (and especially /var/log) on a partition that can be filled by an individual who has access to your system with only basic user permissions. For example, having /var on the same disk partition as /home and not having quota restrictions in place. Additional content can be found regarding what will happen to applications if application/service critical directories such as /tmp, /var/tmp, /var/spool/mail, /var/postgres, etc., can easily be filled by even a non-malicious user. Many default disk partitioning plans, including those of RedHat and many others, create a very small '/' partition. So, it can quickly be filled before you even notice that you've had a problem. Some distributions will also not properly separate critical portions of the file system onto independent partitions (such as /var and /home). So, I'd recommend that while you should do what ever works for you, it would also be good to think about what could possibly go wrong. I've run into problems myself before on Gentoo by not creating a separate partition for /afs and then having an AFS boot sequence fail because the local AFS service was having problems. I then, stupidly, had the same problem some time later on a RH system by not creating a separate LV for /mnt and then having a SAN array mount failure during boot. - VAB - V. Alex Brennen [EMAIL PROTECTED] Senior UNIX Systems Administrator MIT Libraries E25-131 x3-9327 http://vab.mit.edu/ ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Please trim quoted text (was: General Procedure)
> Date: Wed, 9 Jul 2008 22:11:29 -0400 > From: "Ben Scott" <[EMAIL PROTECTED]> > People: When sending replies -- especially one-line replies -- > please trim the hundreds or thousands of lines of quoted text which > add nothing to the context of the message. The server actually Another thing which helps decrease "noise" in list messages is turning off HTML mail. I'm fairly confident an audience of Linux enthusiasts can read text-only email. When a message contains a text/html part, it ends up being more than twice the size. This is especially true if the HTML includes lots of style definitions. ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Please trim quoted text (was: General Procedure)
On Thu, Jul 10, 2008 at 11:07 AM, David Berube <[EMAIL PROTECTED]> wrote: > FWIW, I'd also recommend the QuoteCollapse ThunderBird extension if > you're on Thunderbird ... GMail has a similar feature. But it's important to realize that the hidden quotes are still there in the message; *your* mail program is just hiding them from your view. -- Ben ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Please trim quoted text (was: General Procedure)
On Thu, Jul 10, 2008 at 11:07 AM, David Berube < [EMAIL PROTECTED]> wrote: > FWIW, I'd also recommend the QuoteCollapse ThunderBird extension if > you're on Thunderbird - > it collapses all qoutes to just the first line by default, and you can > click a little plus icon to expand them if you need the full details. Not to nitpick, but we will still have to trim quotes. Quote collapsing is great for the readers, but it won't help with message size limits on the server :-) ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
SL5.2 (RHEL5.2) Print Notifier (cups?)
After upgrading to 5.2 my printing is broken. Print Notifier is not allowing cancellation of jobs. CUPS shows my default printer (which worked last week) with the following "Unable to lookup up host '' - unknown host" I cannot cancel the jobs in cups and see anything show up in print notifier. I have not been able to print since upgrading. Is there a way to clear all pending jobs in print notifier? These jobs even survive a reboot. This appeared to be a problem in FC3 - at least my googling found that. The last time I had issues with printing it was related to having to append a corporate nameserver onto the end of some list. I think I ended up creating a file dhclient-enter-hooks with an export search. Well that file is still there. Any other place to look? ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Adding a new drive / fstab
On Wed, Jul 09, 2008 at 11:50:46AM -0400, Labitt, Bruce wrote: > In the endless pursuit of upgrading this machine I have added a hard > drive to my computer. I have used fdisk to create a linux partition to > the whole disk. I made the disk use the ext3 file system. > > So now for fstab. What is the philosophy for creating an entry? At > this point I'm not sure what the mount point should be. /home sounds I generally mount additional "disks" for data storage as /data/n/. Simple, clean, easy to understand & maintain. These days, for production servers, my typical partitioning is: /boot / /tmp /var /opt (if I'm going to install much 3rd party software that will use it) /data /home and /usr/local are symlinks into /data, as is /opt if I haven't given it a separate partition. If additional data "disks" are needed, I either use LVM to add space to /data, or mount the additional volumes as /data/1/, /data/2/, etc. Ocassionally (usually for Oracle servers) I will create other application-specific filesystems, but it is getting more and more rare as app developers gain clue. -- [EMAIL PROTECTED] OpenPGP KeyID 0x57C3430B Holder of Past Knowledge CS, O- "Would you rather me stick it in your chicken?" Richard Jerrell ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Please trim quoted text (was: General Procedure)
FWIW, I'd also recommend the QuoteCollapse ThunderBird extension if you're on Thunderbird - it collapses all qoutes to just the first line by default, and you can click a little plus icon to expand them if you need the full details. Take it easy, David Berube ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Please trim quoted text (was: General Procedure)
Thank you, Ben, for speaking up. It sucks to have to play cop/babysitter but things were getting totally out of hand here and our normally good S/N ratio is definitely worth defending. ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Please trim quoted text (was: General Procedure)
> People: When sending replies -- especially one-line replies -- > please trim the hundreds or thousands of lines of quoted text which > add nothing to the context of the message. The server actually > started rejecting messages in this thread because they had exceeded > the 40 kilobyte message The server was also sending out several digests with a single message. I thought someone had changed the configuration until I looked at one. -Ric Werme ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
RE: Please trim quoted text (was: General Procedure)
Sorry to the list. I'll try to be more considerate. Regards, Bruce From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of kenta Sent: Thursday, July 10, 2008 8:33 AM To: Greater NH Linux User Group Subject: Re: Please trim quoted text (was: General Procedure) I AGREE. :D :D :D /me ducks On 7/9/08, Ben Scott <[EMAIL PROTECTED]> wrote: On Wed, Jul 9, 2008 at 1:30 PM, Labitt, Bruce <[EMAIL PROTECTED]> wrote: > How does one change that? Sorry to be such a noob. [followed by tons of quoted text] People: When sending replies -- especially one-line replies -- please trim the hundreds or thousands of lines of quoted text which add nothing to the context of the message. The server actually started rejecting messages in this thread because they had exceeded the 40 kilobyte message size limit. Thanks. -- Ben ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/ ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
[GNHLUG] MerriLUG Nashua, Thur 17 Jul, Open Source Advocacy - Want to Help?
Who : Mark Boyajian, IT Consultant, Simple Solutions What : Open Source Advocacy - Want to Help? Where: Martha's Exchange Day : Thur 17 Jul **Next Week** Time : 6:00 PM for grub, 7:30 PM for discussion (usually upstairs) :: Overview Open Source software can be an easy path to Linux. If the Windows user is leary about changing operating systems and software at the same time, high quality Open Source software can make it a much less daunting two step process. Mark Boyajian has a campaign to publicize the Open Source option and inform people about the many fine programs available. His approach is particularly innovative: use the local public library as a base. Mark will explain the very successful Tech Talk series he pioneered at the Lawrence Library in Pepperell along with other strategies. Want to ease your neighbor, cousin, or boss into Linux? Use the two step Open Source route. Take them to Perrerell or, better yet, get your local public library involved! Mark will get you started. >>> RSVP to Jim Kuzdrall for dinner to assure adequate seating. <<< !!! If you are not a "Regular Attendee" (50%), please let me know. !!! Driving directions: http://wiki.gnhlug.org/twiki2/bin/view/Www/PlaceMarthasExchange Thanks, Jim Kuzdrall [EMAIL PROTECTED] ___ gnhlug-announce mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-announce/ ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
Re: Please trim quoted text (was: General Procedure)
I AGREE. :D :D :D /me ducks On 7/9/08, Ben Scott <[EMAIL PROTECTED]> wrote: > > > > On Wed, Jul 9, 2008 at 1:30 PM, Labitt, Bruce > <[EMAIL PROTECTED]> wrote: > > How does one change that? Sorry to be such a noob. > [followed by tons of quoted text] > > People: When sending replies -- especially one-line replies -- > please trim the hundreds or thousands of lines of quoted text which > add nothing to the context of the message. The server actually > started rejecting messages in this thread because they had exceeded > the 40 kilobyte message size limit. > > Thanks. > > > > -- Ben > ___ > gnhlug-discuss mailing list > gnhlug-discuss@mail.gnhlug.org > http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/ > ___ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
