[SLUG] apt-proxy-v2
hi guys i'm trying to use apt-proxy-v2 but it fails to start. # /etc/init.d/apt-proxy-v2 start Starting apt-proxy-v2 Failed to load application: No module named apt_pkg i'm not familiar with python and how it handles lib paths. it may be that on my system i have python2.2 and python2.3. apt-proxy uses 2.2 but apt_pkg is in the 2.3 lib tree. i tried putting apt_pkg into the 2.2 lib tree but it made no difference. a google search just seems to show up posts of people having a problem with apt-listchanges. ideas? marty -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] pppd+mschap-v2 patch
On Tue, 2003-11-18 at 17:25, Chris Barnes wrote: Does anyone know where I can find patches to patch pppd to work with mschap-v2? Hey hey. MS-CHAP support is built in to pppd as of 2.4.2, but I'm not sure if distributions are enabling it in their builds (I know it's not present in my debian package..) So, yeah. Just grab the newest tarball from ftp://ftp.samba.org/pub/ppp and go from there. Be sure to check the docs, though, because some of the option names have been changed to fit in with the rest of the pppd options. -- Pete -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] ssh error message
I have a problem with ssh when I use it to log onto some (but not all) remote hosts. I am assuming the error is coming from ssh. On some machines I get an error message:- Address family not supported by protocol however the logon goes ahead anyway. Anybody got any suggestions about this ? cheers, Bernard Doyle -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] getting root's bin paths after su ?
On Tue, 18 Nov 2003 [EMAIL PROTECTED] wrote: I've just ssh'ed to my server, then , su, and.. hmmm, none of my commands that I normally use work anymore: # service bash: service: command not found after I recovered from the brief hart failure, I realized that service and others still are there, once I prepend the full path so, id I ssh as 'voytek' the su, is there a way to aquire same environment I get as root ? Try su - and entering the root password. DaZZa -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
RE: [SLUG] Questions about AWStats Logs !!
Hi Joel: I have read the manual and also found the config file to modify. I have some questions see below under [Louis]. Anyone else, feel free to answer them as well. -Original Message- From: Joel Heenan [mailto:[EMAIL PROTECTED] Sent: Wednesday, 29 October 2003 09:34 To: Net Business International; [EMAIL PROTECTED] Subject: RE: [SLUG] Questions about AWStats Logs !! Issue 4: == Cron scripts that accesses web pages not via browser but command line. Does AWStats count stats for this as hits and logged them as Unique Visitors for host that resolve to the domain name where the cron is executed ? By default they are included. There is an option called SkipHosts which can be used to skip them. [Louis] I see. What conf file do I have to modify ? For SkipHost I just enter the full IP address of the server ? Also can I use this SkipHost so that when I connect to the site from my dial up ISP AWStats won't log these hits ? http://awstats.sourceforge.net/docs/index.html This explains where the config file should be and what it is called. I don't mean to shrug off your question but you are honestly better off reading the official manual for this. I will only explain it more poorly. Have a look at the documentation in the conf file and on the website about SkipHosts. You should not need to enter the full IP of the server because it will only access itself through localhost. To ignore all hits from your ISP you may need to ignore everyone going through your ISP which may or may not be a good idea. If you are on a major ISP then your reporting may become skewed. [Louis] My objective here is first to have AWStats stop counting hits fron cron jobs. For each web site on the server, do I enter the IP address of the server itself in each site AWStats config file or do I enter the IP address of the site hostname in SkipHosts in each site AWStats config file? [Louis] Most of the hits from the Hosts (Top 10) section of AWStats is from my ISP as a result of the activities I do regularly on the server. I don't want AWStats to look at my ISP IP period. The ISP IP is dynamically allocated. So do I just enter say the top 2 level only, i.e xxx.xxx.ignore.ignore Where xxx.xxx is the part I enter ? [Louis] I have other questions from AWStats log display. With the Connect to site from section of AWStats: 1. the part Links from an Internet Search Engine, the data is provided in two parts, i.e - engine Pages/Hits 1.1- What does Hits mean here ? Does it mean the no. of times Search Engine spidered the site ? 1.2- What does pages viewed means here ? Does it mean total pages that the Search Engine spidered on the site ? 2. the part Links from an external page, sometimes I see pages from the site itself listed in this section. How can that be ? Thanks. -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] ssh error message
On Tue, Nov 18, 2003, Bernard Doyle wrote: I have a problem with ssh when I use it to log onto some (but not all) remote hosts. I am assuming the error is coming from ssh. On some machines I get an error message:- Address family not supported by protocol however the logon goes ahead anyway. Anybody got any suggestions about this ? A quick Google search (Address family not supported by protocol ssh) suggests that it is an error that sometimes happens when X11 forwarding over ssh is turned on. SSH can't forward X11 data -- but this still leaves ssh usable. Check the results of that Google query for more info. -Mary -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Asterisk VoIP usage and technical questions for australian conditions!
I sent an email to the Asterisk mailling list and got few and varied responses. I would be intrested in finding out if there are SLUG members using or deploying asterisk in australia? Would also be intrested in sucess/failure stories on deployment of ISDN cards to run with asterisk? Just looking at setting up my small office phone system for WAN type use so VoIP is what best suits my needs! If there are people who I have not already spoken to who could help me with their info that would be great. Thanks in advance for any help. David Uzzell -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Power supply for router
I have a D-Link router, and a US power supply. Does anyone know of a supplier of power supplies for Australia? It must plug into the wall (240v) and put out 5v at 2.5A (according to the label on the existing power supply. Alternatively a step down transformer for less than the $100 Dick Smith wants. Cheers, Bret -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Power supply for router
In Sydney http://www.wescomponents.com/ should be able to help you with what ever you need! They will ship as well David Uzzell Bret Comstock Waldow wrote: I have a D-Link router, and a US power supply. Does anyone know of a supplier of power supplies for Australia? It must plug into the wall (240v) and put out 5v at 2.5A (according to the label on the existing power supply. Alternatively a step down transformer for less than the $100 Dick Smith wants. Cheers, Bret -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
RE: [SLUG] Questions about AWStats Logs !!
-Original Message- From: Louis [mailto:[EMAIL PROTECTED] Sent: Tuesday, November 18, 2003 9:21 PM To: [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: RE: [SLUG] Questions about AWStats Logs !! Hi Joel: I have read the manual and also found the config file to modify. I have some questions see below under [Louis]. Anyone else, feel free to answer them as well. [Louis] My objective here is first to have AWStats stop counting hits fron cron jobs. For each web site on the server, do I enter the IP address of the server itself in each site AWStats config file or do I enter the IP address of the site hostname in SkipHosts in each site AWStats config file? If cron is running on localhost, you enter localhost. [Louis] Most of the hits from the Hosts (Top 10) section of AWStats is from my ISP as a result of the activities I do regularly on the server. I don't want AWStats to look at my ISP IP period. The ISP IP is dynamically allocated. So do I just enter say the top 2 level only, i.e xxx.xxx.ignore.ignore Where xxx.xxx is the part I enter ? You enter ^xxx.xxx. as in the example directly above SkipHosts. [Louis] I have other questions from AWStats log display. With the Connect to site from section of AWStats: 1. the part Links from an Internet Search Engine, the data is provided in two parts, i.e - engine Pages/Hits 1.1- What does Hits mean here ? Does it mean the no. of times Search Engine spidered the site ? I think hits is files, number of things accessed. So if your webpage has an external stylesheet, or some external javascript, the spider may grab these. Google may even get the images for its image search. 1.2- What does pages viewed means here ? Does it mean total pages that the Search Engine spidered on the site ? It means the number of pages viewed, the number of .php, .html, .htm, etc. files accessed I guess. 2. the part Links from an external page, sometimes I see pages from the site itself listed in this section. How can that be ? Not sure, that is a bit perplexing. Do they have the same hostname? Make sure your HostAliases is set up correctly if you have missed a host here that your site could be called then it will think its external. Thanks. No worries hope that helps. Joel -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Memory Usage
Hi all, Below is part of the 'top' display from one of my servers. Note that it is sorted on Memory and that the top few use only 1.1% of memory, yet the summary at the top would indicate that about 90% or more of memory was in use. Can anyone give me some possible reasons / fixes for this or am I just reading it wrong? 7:47am up 22 days, 18:29, 1 user, load average: 0.12, 0.07, 0.01 139 processes: 138 sleeping, 1 running, 0 zombie, 0 stopped CPU states: 0.7% user, 0.3% system, 0.0% nice, 98.8% idle Mem: 771132K av, 760852K used, 10280K free,5740K shrd, 222472K buff Swap: 522072K av,5992K used, 516080K free 166472K cached PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND 29455 nlazo 9 0 8868 8868 1036 S 0.0 1.1 0:00 aucemail 1004 mysql 9 0 8164 8164 2312 S 0.0 1.0 0:27 mysqld 1018 mysql 8 0 8164 8164 2312 S 0.0 1.0 0:21 mysqld 1019 mysql 9 0 8164 8164 2312 S 0.0 1.0 2:38 mysqld 1020 mysql 9 0 8164 8164 2312 S 0.0 1.0 0:00 mysqld 27344 apache 9 0 7488 7488 7000 S 0.0 0.9 0:00 httpd 27345 apache 9 0 7488 7488 7000 S 0.0 0.9 0:00 httpd 27338 apache 9 0 7384 7384 7000 S 0.0 0.9 0:00 httpd 27341 apache 8 0 7384 7384 7000 S 0.0 0.9 0:00 httpd 27343 apache 8 0 7384 7384 7000 S 0.0 0.9 0:00 httpd 27337 apache 9 0 7380 7380 7032 S 0.0 0.9 0:00 httpd 27342 apache 9 0 7380 7380 7000 S 0.0 0.9 0:00 httpd 27346 apache 9 0 7380 7380 7000 S 0.0 0.9 0:00 httpd 28333 apache 9 0 7380 7380 7028 S 0.0 0.9 0:00 httpd -- Simon Bryan IT Manager OLMC Parramatta -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Power supply for router
On Wed, Nov 19, 2003 at 12:32:10AM +1100, Bret Comstock Waldow wrote: I have a D-Link router, and a US power supply. Does anyone know of a supplier of power supplies for Australia? I got a good one from jaycar, put MF1091 in the keyword search at www.jaycar.com.au It's $49.95 -- I don't think you'll find anything cheaper than that unless you try ebay, etc. -i [EMAIL PROTECTED] http://www.gelato.unsw.edu.au -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Teraterm and control-c
Peter Vogel wrote: I'm using Teraterm under Windows as a terminal to a linux box. After much we searching I still can't figure out how to make Teraterm send a controlC (to break). Any pointers appreciated. Hi Peter, I think you might find that Alt-B sends a break on teraterm. I used to use it but switched to Putty a while ago. Cheers, Brad -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Linux FileServer Mac Clients permission problems
Sluggers, Not knowing a thing about Apple Macs I've been asked to look at a problem where a Linux Box is acting as a file server to a bunch of Macs. For some files, when the user on the Mac tries to open a file (in Word) a dialog comes up stating that the file is being modified by another user (but it isn't). The files in question appear to have different ownership and group (i.e rw-rw-r) to files that don't cause this problem. But changing the linux permissions on the file server makes no difference. The Linux box is running appletalk and Samba. The users claim that it happens whether they use appletalk or tcp/ip. (obviously tcp/ip is too low a level protocol so I figure that they mean smb but refer to caveat about my knowledge of Macs). Is there some sort of semaphore system in use here? I.E does the macs keep track of who has what file open in some other manner? Thanks for any pointers Pete. -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Asterisk VoIP usage and technical questions for australian conditions!
David Uzzell wrote: I sent an email to the Asterisk mailling list and got few and varied responses. I would be intrested in finding out if there are SLUG members using or deploying asterisk in australia? No, but I'm about to. Would also be intrested in sucess/failure stories on deployment of ISDN cards to run with asterisk? The one you're supposed to use is the new one from Digium, the TE410P. Apparently it works fine in Australia. It may be overkill if you don't need 4 ISDN ports but it works and it's not super expensive(*). Theoretically the E100P should work as well but I've heard of problems with it. This is just of the various Asterisk mailing lists. (*) In comparison, I have a quote from a cisco dealer for their call manager hardware/software on my desk. It's astounding. Just looking at setting up my small office phone system for WAN type use so VoIP is what best suits my needs! Let me know how you go. -- Del -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] ssh error message
On Tue, Nov 18, 2003 at 09:31:28PM +1100, Mary Gardiner wrote: On some machines I get an error message:- Address family not supported by protocol however the logon goes ahead anyway. Anybody got any suggestions about this ? It happens to me when the destination has an IPv6 address and your machine doesn't do v6. -- Christopher Vance -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Memory Usage
Simon Bryan wrote: Hi all, Below is part of the 'top' display from one of my servers. Note that it is sorted on Memory and that the top few use only 1.1% of memory, yet the summary at the top would indicate that about 90% or more of memory was in use. Can anyone give me some possible reasons / fixes for this or am I just reading it wrong? 7:47am up 22 days, 18:29, 1 user, load average: 0.12, 0.07, 0.01 139 processes: 138 sleeping, 1 running, 0 zombie, 0 stopped CPU states: 0.7% user, 0.3% system, 0.0% nice, 98.8% idle Mem: 771132K av, 760852K used, 10280K free,5740K shrd, 222472K buff Swap: 522072K av,5992K used, 516080K free 166472K cached PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND 29455 nlazo 9 0 8868 8868 1036 S 0.0 1.1 0:00 aucemail 1004 mysql 9 0 8164 8164 2312 S 0.0 1.0 0:27 mysqld 1018 mysql 8 0 8164 8164 2312 S 0.0 1.0 0:21 mysqld 1019 mysql 9 0 8164 8164 2312 S 0.0 1.0 2:38 mysqld 1020 mysql 9 0 8164 8164 2312 S 0.0 1.0 0:00 mysqld 27344 apache 9 0 7488 7488 7000 S 0.0 0.9 0:00 httpd 27345 apache 9 0 7488 7488 7000 S 0.0 0.9 0:00 httpd 27338 apache 9 0 7384 7384 7000 S 0.0 0.9 0:00 httpd 27341 apache 8 0 7384 7384 7000 S 0.0 0.9 0:00 httpd 27343 apache 8 0 7384 7384 7000 S 0.0 0.9 0:00 httpd 27337 apache 9 0 7380 7380 7032 S 0.0 0.9 0:00 httpd 27342 apache 9 0 7380 7380 7000 S 0.0 0.9 0:00 httpd 27346 apache 9 0 7380 7380 7000 S 0.0 0.9 0:00 httpd 28333 apache 9 0 7380 7380 7028 S 0.0 0.9 0:00 httpd This looks pretty normal. Notice the buff and cache are about 217Mb and 163Mb respectively? The Linux kernel will try and use all the RAM is can for buffers and cache, after accomodating the programs etc, to minimise disk/network access (which is much slower than RAM - naturally). However, if a program needed more space (heap/stack etc) the kernel will sacrifice buff and/or cache to make room. The only time to worry is when you see a lot of swap in use and large amount of paging activity. This is usually a good sign that either something has sprung a leak, or that you need more RAM. However, if a program starts then goes idle for an extended period, the kernel may well swap it out of RAM to either make space for another program or allocate more buffers/cache. I'm not that familiar with the actual complexities of the kernel's memory allocation algorithms but you get the general idea. HTH. Cheers, James -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Asterisk VoIP usage and technical questions for australian conditions!
On Wed, 2003-11-19 at 09:34, Del wrote: David Uzzell wrote: I sent an email to the Asterisk mailling list and got few and varied responses. I would be intrested in finding out if there are SLUG members using or deploying asterisk in australia? No, but I'm about to. Me too. Would also be intrested in sucess/failure stories on deployment of ISDN cards to run with asterisk? The one you're supposed to use is the new one from Digium, the TE410P. Apparently it works fine in Australia. It may be overkill if you don't need 4 ISDN ports but it works and it's not super expensive(*). Theoretically the E100P should work as well but I've heard of problems with it. This is just of the various Asterisk mailing lists. I've heard success stories also with the AVM!FritzCard (Austel Approved) which is worth approximately $350 (last time I was quoted). I've been told (check the Asterisk mailing list archives) that the NetJet ISDN cards (about half the price of the AVM!FritzCard) have some echo problems and don't have CAPI support, so I'm leaning towards getting the AVM. As for phones, I'm looking at getting a hardware SIP phone from http://www.chagres.net/products/voip/phones.html or I'll just use the X-Lite/X-Pro client from Xten (www.xten.com). I've been researching and testing Asterisk for the last week so the above is just some outcome of the research I conducted. HTH. Regards, Gonzalo -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Memory Usage
On Tue, 2003-11-18 at 21:42, Simon Bryan wrote: Hi all, Below is part of the 'top' display from one of my servers. Note that it is sorted on Memory and that the top few use only 1.1% of memory, yet the summary at the top would indicate that about 90% or more of memory was in use. Nothing really more to add than what James said - Linux caches disk activity with spare RAM, and keeps a little around to hand out on the fly should it need to. Other than to note that worrying about free ram is kinda, um, Windowsish. You paid for the RAM, of course you want it to be used! Mike -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Change email address
HI All wondering if someone could point me in the right direction to change my email address on here Cheers Trevor -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Linux FileServer Mac Clients permission problems
On Wed, 19 Nov 2003, Peter Rundle wrote: Sluggers, Not knowing a thing about Apple Macs I've been asked to look at a problem where a Linux Box is acting as a file server to a bunch of Macs. For some files, when the user on the Mac tries to open a file (in Word) a dialog comes up stating that the file is being modified by another user (but it isn't). Can't help with this one.. i've not experienced it The files in question appear to have different ownership and group (i.e rw-rw-r) to files that don't cause this problem. But changing the linux permissions on the file server makes no difference. The Linux box is running appletalk and Samba. The users claim that it happens whether they use appletalk or tcp/ip. (obviously tcp/ip is too low a level protocol so I figure that they mean smb but refer to caveat about my knowledge of Macs). I don't think Mac's talk smb at all, although I could be wrong since I happily never use windows :) You can use appletalk over a local network, or appletalk over tcp/ip. I'm definitely NOT a network engineer, but I think appletalk becomes another layer within tcp/ip. It's an option in the appletalk control panel in older mac's. I use this facility all the time, and as far as the mac's are concerned, once you connect via tcp/ip it just looks like appletalk. Older mac's have a permission system vaguely similar to unix, while OS X is unix, but I have no idea why the files should be appear to be in use, unless they actually are! How many mac's on the system? Perhaps someone else doesn't ever get around to closing files.. i have a number of people on my system who have no idea about quitting or closing anything ever. When I have this problem, that's always been the cause. Is there some sort of semaphore system in use here? I.E does the macs keep track of who has what file open in some other manner? Thanks for any pointers Pete. -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Change email address
On Wed, Nov 19, 2003, Trevor wrote: HI All wondering if someone could point me in the right direction to change my email address on here Go to http://lists.slug.org.au/options/slug/trevor--at--txc.net.au and change it using the Changing your slug membership information section of that page. -Mary -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Change email address
Trevor wrote:
HI All
wondering if someone could point me in the right direction to change my
email address on here
...snip
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
The link above should tell you how unsubscribe your old address and
subscribe your new address.
--
Terry Collins {:-)}}} email: terryc at woa.com.au www:
http://www.woa.com.au
Wombat Outdoor Adventures Bicycles, Computers, GIS, Printing,
Publishing
People without trees are like fish without clean water
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Memory Usage
James Gray said: Simon Bryan wrote: Hi all, Below is part of the 'top' display from one of my servers. Note that it is sorted on Memory and that the top few use only 1.1% of memory, yet the summary at the top would indicate that about 90% or more of memory was in use. Can anyone give me some possible reasons / fixes for this or am I just reading it wrong? 7:47am up 22 days, 18:29, 1 user, load average: 0.12, 0.07, 0.01 139 processes: 138 sleeping, 1 running, 0 zombie, 0 stopped CPU states: 0.7% user, 0.3% system, 0.0% nice, 98.8% idle Mem: 771132K av, 760852K used, 10280K free,5740K shrd, 222472K buff Swap: 522072K av,5992K used, 516080K free 166472K cached This looks pretty normal. Notice the buff and cache are about 217Mb and 163Mb respectively? The Linux kernel will try and use all the RAM is can for buffers and cache, after accomodating the programs etc, to minimise disk/network access (which is much slower than RAM - naturally). However, if a program needed more space (heap/stack etc) the kernel will sacrifice buff and/or cache to make room. The only time to worry is when you see a lot of swap in use and large amount of paging activity. This is usually a good sign that either something has sprung a leak, or that you need more RAM. However, if a program starts then goes idle for an extended period, the kernel may well swap it out of RAM to either make space for another program or allocate more buffers/cache. I'm not that familiar with the actual complexities of the kernel's memory allocation algorithms but you get the general idea. OK, thanks for that, I note in the reply from Mike the comment about 'windowish' worrying about RAM usage - I had just finished looking through my WIndows servers memory requirements, which is a different kettle of fish to this. I then noticed that all my Linux servers were in a similar state to this (except one where the virus scanner had gone psycho and never finished - over 5 days 5 different instances of it were still running (the scan shoud take about 10 minutes at 1am in the morning). Killing them off and reconfiguring the virus scanner freed a lot of memory and CPU! Good to keep learning about these things. Cheers -- Simon Bryan IT Manager OLMC Parramatta -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Memory Usage
On Wed, 2003-11-19 at 03:06, Simon Bryan wrote: James Gray said: OK, thanks for that, I note in the reply from Mike the comment about 'windowish' worrying about RAM usage More but worrying about not having too much free. The usage of individual apps though matters. Mike -- __ Mike MacCana ConsultantRHCX, MCSE, MCP+I Cybersource: Providing Quality IT Professional Services for 11 Years Specialists in Unix/Linux, TCP/IP and Web Application Development Level 4, 10 Queen St, Melbourne. Ph : 03 9621 2377 Fax: 03 9621 2477 -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] getting root's bin paths after su ?
A little email about .bashrc vs .bash_profile: On Tue, 2003-11-18 at 15:28, Andrew Bennetts wrote: The dash makes it a login shell, which means it executes the bashrcs and whatnot specific to the root user, thus you'll get the environment you expect. Without wanting to be pedantic, I got this wrong enough times to offer a minor clarification: A great way to figure out what your system is doing is to put echo In ~andrew/.bashrc 12 As a line in your .bashrc, and similar lines in places like .bash_profile, .profile, and if you're really stumped, /etc/profile, .gnomerc and /etc/X11/gdm/gnomerc (but see warning [1] below!) Sometimes you'll be surprised. For example, I was once dealing with Solaris boxes whose default shell wasn't bash. I wanted to use bash, and so would do ssh [EMAIL PROTECTED] bash [assuming bash was even on default path, groan] The thing that surprised me was that, *I* thought I was logging in, but in this case, .bash_profile wasn't run, only .bashrc . This because running bash as the command via ssh is the same as if you just type `bash` on a command line - .bashrc gets sourced, but not .bash_profile. shrug, whatever, fair enough. As you may have noticed, one of bash's options is --login (this is what relates to the `su -` conversation). It turned out that if I wanted .bash_profile to run, I needed a login environment and so needed to amend the above command to ssh [EMAIL PROTECTED] 'bash --login' which was a bit of a pain. [So I wrote a wrapper, but that's another story ;)] So this is all to say that, especially with bash as a shell (but I suspect many others), it pays to a) put environment setting/creating things, like setting PATH, in .bashrc b) put login only things, like checking new mail or changing directory to ~ or running a clock or ... in .bash_profile; c) go to the effort to actually *get* a login shell when you want one. Note: Interesting how the default skeleton .bash_profile always sources .bashrc for you! That's because if bash IS in login mode, and decides to source .bash_profile, it DOESN'T source .bashrc automatically as well - so if you're environment settings are there, then you need to act to get them. -- [1] WARNING! Don't forget that since .bashrc gets sourced any time a shell runs, it REALLY must not output anything on stdout - in other words, that little debuggy statement I suggested above will wreck your world if it *doesn't* go to stderr and you try to do `scp`. So don't forget 12 on any debug lines Ah, the things we learn by trial and error. Given that this is all open source and subject to modification without notice, your mileage may vary. Certain forward looking statements (like it may work) are not to be taken as authoritative. And if this message wasn't intended for you, then destroy it without reading it :) AfC -- Andrew Frederick Cowie Operational Dynamics Consulting Pty Ltd Australia: +61 2 9977 6866 http://www.operationaldynamics.com/ -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] ntp/ntpdate no suitable server ?
** Reply to note from John Clarke [EMAIL PROTECTED] Tue, 18 Nov 2003 11:36:15 +1100 John, For ntpd, I'd limit incoming packets to the ip address(es) of your time server(s) and source and destination ports 123. For ntpdate, you'll need to allow source port 123 and destination ports 1024:65535, but still limit it to the address(es) of your time server(s). are all of the above in: ipchains ? ntp.conf ? Voytek Eymont -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] getting root's bin paths after su ?
On Wed, Nov 19, 2003 at 02:25:33PM +1100, Andrew Cowie wrote: [1] WARNING! Don't forget that since .bashrc gets sourced any time a shell runs, it REALLY must not output anything on stdout - in other words, that little debuggy statement I suggested above will wreck your world if it *doesn't* go to stderr and you try to do `scp`. So don't forget 12 on any debug lines Hopefully this particular bit of pain will end soon as recent openssh's use the sftp protocol when scp'ing. Matt -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] ntp/ntpdate no suitable server ?
On Wed, Nov 19, 2003 at 02:53:12 +, Voytek wrote: For ntpd, I'd limit incoming packets to the ip address(es) of your time server(s) and source and destination ports 123. For ntpdate, you'll need to allow source port 123 and destination ports 1024:65535, but still limit it to the address(es) of your time server(s). are all of the above in: ipchains ? ntp.conf ? ipchains/iptables, but you should also restrict the level of access granted to your time server in ntp.conf. There was a good example posted a couple of weeks ago which you can find in the list archives. Cheers, John -- whois [EMAIL PROTECTED] GPG key id: 0xD59C360F http://kirriwa.net/john/ -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Samba Problems
Hi All, Having a few issues with a samba share that i cannot work out... Basically i can access a share called shared on my samba server but nobody else can do it on their machines, even though they are part of the same group and that group has access to the share - otherwise i myself wouldnt be able to access it... The other users seem to be authenticating, because they can see their home directory and write/read to it/from it... here is the smb.conf: [global] server string = Samba Server %v security = user map to guest = bad user workgroup = WG1 load printers = yes log file = /var/log/samba/log.%m smb passwd file = /etc/samba/smbpasswd dns proxy = no printcap name = cups encrypt passwords = yes wins support = true max log size = 50 printer admin = @adm socket options = TCP_NODELAY SO_RCVBUF=8192 SO_SNDBUF=8192 printing = cups unix password sync = yes netbios name = woja os level = 20 [homes] comment = Home Directories browseable = no writable = yes [shared] writeable = yes valid users = @shared user = @shared path = /mnt/shared comment = shared folder /mnt/shared is owned by the group shared and all have read write permissions recursively.. Each time i try to browse the shared share on one of the other workstations, this appears in the appropriate log file: [2003/11/19 14:23:45, 0] smbd/service.c:make_connection(599) compaq (192.168.1.102) Can't change directory to /mnt/shared (Permission denied) [2003/11/19 14:23:45, 0] smbd/service.c:make_connection(599) compaq (192.168.1.102) Can't change directory to /mnt/shared (Permission denied) where else can their permission be denied ??? it doesnt make sense... Thanks for any help.. Adam W. -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Continuation of PPP woes.
G'day all... I'm still having some PPP woes. PPP doesn't always reconnect properly upon reboot - which then causes squid to fall-over - and sometimes PPP falls over by itself and doesn't manage to reconnect. If it was just for me, it'd be ok, as I can fix it - but its for an organisation I volunteer for, so it causes drama when I'm not around and they can't access the internet. I need to create a solution so that it automagically corrects itself, or the fault is prevented from happening. At this stage I'm not sure where the fault lies. (Eg. The line itself, SwiftDSL or our server.) I tried putting MAXFAIL=0 into the options file for the device which someone else suggested. (Sorry, can't remember who, but thanks for their help.) Unfortunately this has only helped marginally, and the problem persists. Attached is a cut down version of the /var/log/messages file from the machine. (IP addresses have been blanked out.) I've tried to diagnose the fault myself, but I'm not really that au-feaux (sp) with PPP - I'd really appreciate a hand on this one. Many thanks and all the best. Mike Kraus mail2web - Check your email from the web at http://mail2web.com/ . Nov 17 12:25:48 rosebud network: Bringing up interface ppp0: failed Nov 17 12:25:58 rosebud pppd[2177]: Connection terminated. Nov 17 12:25:58 rosebud pppd[2177]: LCP: timeout sending Config-Requests Nov 17 12:26:02 rosebud pppd[2177]: Exit. Nov 17 12:26:02 rosebud pppoe[2178]: Timeout waiting for PADS packets Nov 17 12:35:01 rosebud pppd[2918]: Connect: ppp0 -- /dev/pts/0 Nov 17 12:35:01 rosebud pppd[2918]: pppd 2.4.1 started by root, uid 0 Nov 17 12:35:01 rosebud pppd[2918]: Using interface ppp0 Nov 17 12:35:02 rosebud /etc/hotplug/net.agent: assuming ppp0 is already up Nov 17 12:35:02 rosebud pppoe[2919]: PPP session is 6178 Nov 17 12:35:05 rosebud pppd[2918]: local IP address ***.***.***.* Nov 17 12:35:05 rosebud pppd[2918]: primary DNS address ***.**.**.** Nov 17 12:35:05 rosebud pppd[2918]: remote IP address ***.***.**.*** Nov 17 12:35:05 rosebud pppd[2918]: secondary DNS address ***.***.**.** Nov 18 16:58:43 rosebud adsl-stop: Killing adsl-connect Nov 18 16:58:43 rosebud adsl-stop: Killing pppd Nov 18 16:58:43 rosebud pppd[2918]: Terminating on signal 15. Nov 18 16:58:44 rosebud pppd[2918]: Connection terminated. Nov 18 16:58:44 rosebud pppd[2918]: Connect time 1703.7 minutes. Nov 18 16:58:44 rosebud pppd[2918]: Sent 2052513 bytes, received 28330500 bytes. Nov 18 16:58:44 rosebud pppoe[2919]: read (asyncReadFromPPP): Session 6178: Input/output error Nov 18 16:58:44 rosebud pppoe[2919]: Sent PADT Nov 18 16:58:45 rosebud pppd[2918]: Exit. Nov 18 16:59:06 rosebud pppd[4481]: Connect: ppp0 -- /dev/pts/0 Nov 18 16:59:06 rosebud pppd[4481]: pppd 2.4.1 started by root, uid 0 Nov 18 16:59:06 rosebud pppd[4481]: Using interface ppp0 Nov 18 16:59:07 rosebud /etc/hotplug/net.agent: assuming ppp0 is already up Nov 18 16:59:07 rosebud pppoe[4482]: PPP session is 6631 Nov 18 16:59:10 rosebud pppd[4481]: local IP address ***.***.***.* Nov 18 16:59:10 rosebud pppd[4481]: primary DNS address ***.***.**.** Nov 18 16:59:10 rosebud pppd[4481]: remote IP address ***.***.**.*** Nov 18 16:59:10 rosebud pppd[4481]: secondary DNS address ***.***.**.** Nov 18 17:05:56 rosebud adsl-connect: ADSL connection lost; attempting re-connection. Nov 18 17:05:56 rosebud adsl-stop: Killing adsl-connect Nov 18 17:05:56 rosebud adsl-stop: Killing pppd Nov 18 17:05:56 rosebud pppd[4481]: Connection terminated. Nov 18 17:05:56 rosebud pppd[4481]: Connect time 6.9 minutes. Nov 18 17:05:56 rosebud pppd[4481]: Exit. Nov 18 17:05:56 rosebud pppd[4481]: Sent 462 bytes, received 378 bytes. Nov 18 17:05:56 rosebud pppd[4481]: Terminating on signal 15. Nov 18 17:05:56 rosebud pppoe[4482]: read (asyncReadFromPPP): Session 6631: Input/output error Nov 18 17:05:56 rosebud pppoe[4482]: Sent PADT Nov 18 17:06:04 rosebud pppd[4788]: pppd 2.4.1 started by root, uid 0 Nov 18 17:06:05 rosebud /etc/hotplug/net.agent: assuming ppp0 is already up Nov 18 17:06:05 rosebud pppd[4788]: Connect: ppp0 -- /dev/pts/0 Nov 18 17:06:05 rosebud pppd[4788]: Using interface ppp0 Nov 18 17:06:05 rosebud pppoe[4789]: PPP session is 6639 Nov 18 17:06:08 rosebud pppd[4788]: local IP address ***.***.***.* Nov 18 17:06:08 rosebud pppd[4788]: primary DNS address ***.***.**.** Nov 18 17:06:08 rosebud pppd[4788]: remote IP address ***.***.**.*** Nov 18 17:06:08 rosebud pppd[4788]: secondary DNS address ***.***.**.** Nov 18 17:08:48 rosebud adsl-stop: Killing adsl-connect Nov 18 17:08:48 rosebud adsl-stop: Killing pppd Nov 18 17:08:48 rosebud network: Shutting down interface ppp0: succeeded Nov 18 17:08:48 rosebud pppd[4788]: Connection terminated. Nov 18 17:08:48 rosebud pppd[4788]: Connect time 2.8 minutes. Nov 18 17:08:48 rosebud pppd[4788]: Exit. Nov 18 17:08:48 rosebud pppd[4788]: Sent 114 bytes, received 114 bytes. Nov 18 17:08:48
[SLUG] Next SLUGAMuSIG meeting - Linux audio/ music
Fourth (fifth?) Meeting of SLUGAMuSIG - Audio/ Music Special Interest Group When: Saturday, November 29th; 10:30am onwards Where: Macquarie University, Department of Contemporary Music Studies, building W6A, room 608 All interested in music/ audio and Linux are invited for the next gathering of the group. The general plan is a demonstration/ discussion in the morning, followed by an open session of jamming/ coding/ talking in the afternoon. All levels of interest are welcomed. Chris Deigan has offered to discuss Gentoo and music apps; additional demos/ discussions welcomed. The best way to get to the Music Department (on the 6th floor of building W6A) is from the Balaclava Rd entrance (opposite Woolies from Epping Rd). The closest parking is W4, and costs $8.00 for the whole day. For people arriving at various times through the day, the front door might be locked, but we'll keep an eye out, or call me on 0408 478 802. Denis Crowdy -- Department of Contemporary Music Studies Macquarie University NSW 2109 Australia, ph: +61 (0)2 9850 6787, fax: 9850 6593 http://www.ccms.mq.edu.au -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Calamaris/Webalizer download count versus Netfilter byte count versus ISP byte count
Hi people, Ive got a real problem here. Ill explain It as best I can. I have set a system and connected it to adsl. All works well. The adsl is shared to about 6 people in the building via Squid cache, and shared to another person via Netfilter/NAT. All works well. I have set the firewall up so that every packet that enters and exits the eth1 (adsl) interface is logged to a mysql database using the ULOG daemon. All works well. Now I use Calamaris or Webalizer to generate a report based on Squid usage so we can see how much people are downloading. It seems to work well. I use PHP scripts to query the MySQL database to find out how much data comes into the eth1 interface. Seems to work fine. Today I requested the current internet usage from our ISP. The problem is that we are getting very conflicting figures in our reports. For the month of november (from the 1st to today the 19th) our ISP tells us we have already downloaded 2.2gigs. Webalizer tells me that we have only downloaded 780megs for November (from 1st to 19th). And when I do a report on the MySQL database it tells me we have only downloaded about 450megs! Its got me completely stumped because everything that comes through eth1 (adsl) is logged to the database, so in theory the figure should be approximate to the figure our ISP gave us. I did a few tests setting up another chain and added some rules to Netfilter, anyway, I reset the firewall counters and downloaded a 512kb file. The web browser reported the file size as slightly larger than 512kb and the byte counters in Netfilter said the data that came through the specific chain I set up for the download was only about 446kb. So can anyone explain why Im getting such varying figures? Are there problems with Netfilters byte counters, or anything like that. Does anyone have any ideas? Very much appreciated. Chris Barnes. -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
