- WLUG - wlug.mailman3.com

Re: [Wlug] LTSP as a means of recycling old PCs...

by Dwight A. Ernest

Bill Smith wrote: > There's a post on Slashdot talking about LTSP... (Yeah, I posted it... <g>) > > The link is http://slashdot.org/article.pl?sid=04/05/22/223425&&mode=thread&tid=126&tid… > > There's one comment that really drives the point home: #9227112 ... > > Anyway, I know that there was a recent WLUG meeting about LTSP, so I thought this might be of interest. I think it was a Boston Linux Users Group meeting that talked about LTSP (my friend Dave Shevett was the presenter, and it's the only such meeting I've attended). I get "Nothing for you to see here. Please move along." at that link. --Dwight

May 23, 2004

LTSP as a means of recycling old PCs...

by Bill Smith

There's a post on Slashdot talking about LTSP... (Yeah, I posted it... <g>) The link is http://slashdot.org/article.pl?sid=04/05/22/223425&&mode=thread&tid=126&tid… There's one comment that really drives the point home: #9227112 ... Anyway, I know that there was a recent WLUG meeting about LTSP, so I thought this might be of interest. -- William Smith wsmith(a)chezsmith.com Fall River, MA

May 23, 2004

Re: [Wlug] Creating web content with Linux

by Steve Freitag

On Wed, 2004-05-05 at 12:12, Andy Stewart wrote: > The questions are: What tools are available on Linux for creating web > content? Is there something akin to FrontPage for Linux? Nvu (http://www.nvu.com) is mentioned on page 10 of the June 2004 Linux Journal. >From the nvu.com website: "A complete Web Authoring System for Linux Desktop users to rival programs like FrontPage and Dreamweaver."

May 22, 2004

mouse scroll wheel

by Gregory Avedissian

I did a new install of SuSE 8.2, and my mouse scroll wheel no longer works. Previous install was on top of 7.2. Current XF86Config mouse section is identical to the one that worked. It also works when I boot into debian. I've tried adding the imwheel command to .xinitrc, but that didn't help. Maybe I don't understand the -W option. What's a GPM fifo? Is that the script file that starts gpm? If it is, I tried that, and it didn't work, either. Also tried adding and removing gpm from run-level services. Any ideas? Thanks, Greg

May 22, 2004

RE: [Wlug] Comcast dolts (mail server problem?)

by Don Levey

wlug-bounces(a)mail.wlug.org wrote: > I sent email to postmaster(a)comcast.net describing the problem. I > received an automated reply about "thank you for reporting network > abuse". This is getting downright comical, in a sad sort of way. They must be so used to reports on Comcast abuse that they use that for an autoresponder on postmaster as well as abuse. I lose track of the number of times I send reports to them about broken Windows installations on t heir cable modems spewing spam and viruses. -Don

May 21, 2004

Re: [Wlug] Re:Comcast dolts (mail server problem?)

by Jeff Kinz

Ignore this message, We are attempting to discern what "suspicious header is causing the filters to "pend" messages for moderator approval. On Fri, May 21, 2004 at 09:20:46AM -0400, Jeff Kinz wrote: > On Fri, May 21, 2004 at 08:52:49AM -0400, Andy Stewart wrote: > > In Kmail, when I configure the email sending parameters, I fill in > > "smtp.comcast.net" as instructed. In the security section, there's a > > selection for "check what the server supports". I do this and it returns > > "encryption TLS, authentication method CRAM-MD5". Well, it does this *most* > > of the time. Once in a while, it returns "encryption SSL". > > > > Keeping the above in mind, I periodically get errors when sending email. The > > error is "authentication failure". However, if I immediately resend, it goes > > through just fine (sometimes after repeated resends). > > > > Note also that smtp.comcast.net maps to 4 different IP addresses. > > > > My theory is that the Comcast servers are not all using the same > > authentication method, and when I catch one that's not setup the same as my > > email program, I don't authenticate and thus cannot send email. Does this > > theory seem reasonable? > > > > I tried to explain my theory to Comcast tech support who asked me to reboot > > Windows, and then complained about my unsupported email client. When I asked > > to speak to the system administrator for the email server, they refused. > > Comcast can't get into their heads that there might be an intelligent user or > > two using their service, and that all problems aren't the fault of the user. > > Sigh...any thoughts ? > > Thoughts beyond what **ses most mass-market broadband providers are ? :) > > I would set up a script to send an email to yourself routed thru an > outside node (I can give you a user a kinz.org that will just forward > stuff back to you). have it periodically emit an email that you can > easily filter out and send to dev null. > > > Set up tcpdump to capture the smtp and auth data. > > See which IP's are using which auth method, (or verify that > there is something else going on... ). > > I use comcast w/sendmail and mutt. All my email is sent out thru > Comcast's smtp service. iMy smtp auth flags are set to "A", whetever > that means. (Can't find it in the bat book! ) > > > -- > Jeff Kinz, Emergent Research, Hudson, MA. > "jkinz(a)kinz.org" is copyright 2004. > Use is restricted. Any use is an acceptance of the offer at > www.kinz.org/policy.html (if you could get there .. ) > > _______________________________________________ > Wlug mailing list > Wlug(a)mail.wlug.org > http://mail.wlug.org/mailman/listinfo/wlug > -- Jeff Kinz, Emergent Research, Hudson, MA. "jkinz(a)kinz.org" is copyright 2004. Use is restricted. Any use is an acceptance of the offer at www.kinz.org/policy.html (if you could get there .. )

May 21, 2004

RE: [Wlug] Comcast dolts (mail server problem?)

by Andy Stewart

I sent email to postmaster(a)comcast.net describing the problem. I received an automated reply about "thank you for reporting network abuse". This is getting downright comical, in a sad sort of way. -- Andy Stewart, Founder Worcester Linux Users' Group Worcester, MA USA http://www.wlug.org

May 21, 2004

Re: [Wlug] GPG - web of trust

by Frank Sweetser

There's a good explanation with more of the nitty gritty details in the online manual at http://www.gnupg.org/gph/en/manual.html#AEN335 -- Frank Sweetser fs at wpi.edu WPI Network Engineer GPG fingerprint = 6174 1257 129E 0D21 D8D4 E8A3 8E39 29E3 E2E8 8CEC

May 21, 2004

Re: [Wlug] GPG - web of trust

by Josh Huber

Frank Sweetser <fs(a)WPI.EDU> writes: > If I know who to look for in between , I can then find a path to the > signature I'm looking for. Unfortuatelly, there's no good way that > I know of to automatically find the shortest trusted path between to > keys. This site is pretty cool: http://www.lysator.liu.se/~jc/wotsap/ For example: http://webware.lysator.liu.se/jc/wotsap/?top=0xE580B363&bottom=0x6B21489A ...and... http://webware.lysator.liu.se/jc/wotsap/?top=0x135EA668&bottom=0x6B21489A Seems like there's at least a small bug -- the email address is shows for me is one of my revoked uids. -- Josh Huber

May 21, 2004

Re: [Wlug] GPG - web of trust

by Frank Sweetser

On Fri, May 21, 2004 at 09:17:04AM -0400, Andy Stewart wrote: > > HI all, > > I recently received an email from Jeff Moyer on the WLUG list. Kmail noted > that the digital signature was valid but untrusted (I thought that feature > was broken - why is it suddenly working??). I had downloaded Jeff's key from > the key server. > > What establishes this key as trusted? Certainly, if I had signed Jeff's key > myself, I would trust it, but in this case, I did not. I also don't know > anybody who has signed Jeff's key. When you sign someone else's public key, your are explicitly stating that you trust that the identity string contained in that pubkey accuratelly identifies the person who created the signature and knows the secret passphrase to the private key. > Do I need to place a personal value on keys which I've downloaded? > Hypothetically, what if I downloaded it from some rogue server masquerading > as the real thing? There's no need to throw a rogue server into the picture. One thing to note about the PGP keyservers - they do absolutelly no authentication at all. Anyone can store any key identifying themselves as anyone at all, and the keyservers will take it. All I have to do is generate a new key pair identifying myself as Jeff Moyer, upload it, and if you search for Jeff on the keyservers, both his and mine will appear. This is the purpose for having key signing parties. By having people physically show up with both their key fingerprints and verifiable evidenice of who they are, you can establish within reasonable doubt that *this* key belongs to this person, regardless of what the identifying strings in the public key advertise. > What I suspect is that if I signed Joe Blow's key, and Joe had signed Jeff > Moyer's key, and I had placed a value on Joe's ability to properly verify > keys, that there would be some implication about the goodness of Jeff's key. > Correct? Correct - this is the "web of trust" in operation. Note that in order for this to work, everyone must have uploaded their signed public keys to the keyservers, and then must periodically refresh their local public keyring from the keyservers to get an up to date list of who has signed who'se keys. gpg only generates its internal web of trust (stored in the trustdb) based off of the local keyring. > In my situation, I suspect that there are too many degrees of separation > between keys which I trust and Jeff's key and its signers. How does this gap > get closed? There are probably missing links, if such a path exists at all. Let's say that you signed Bob's key, Bob signed Charlie's key, and Charlie signed Jeff's key. Since you signed Bob's key, you presumablly have declared it trusted. If you haven't downloaded Charlie's key, though, then gpg can't find a path to Jeff, so it will declare his key as untrusted. To try and close this, you must first find out what other keys have signed the signature you're trying to get to. I'll use Theo's key as an example here (with lots of snipping, since he's got a big keyring). tty/10 1002 erwin 09:34:56 $ gpg --list-sigs felicity pub 1024D/E580B363 1997-11-09 Theo Van Dinter <felicity(a)kluge.net> sig F74F343D 2003-11-01 [User id not found] sig C52DCE75 2003-11-01 [User id not found] sig 3 R CCE3BD36 2003-08-23 [User id not found] sig 3 E2E88CEC 2004-04-23 Frank Sweetser <fs(a)wpi.edu> sig 898AA63C 2004-04-23 Doug Mildram (Born under 4.2BSD) <doug.mildram(a)mindspeed.com> These are a few of the keys in Theo's public keyring, which I have a copy of from the public keyservers. I can see my own in there, and Doug's, since my keyring contains keys corresponding to those fingerprints. The ones marked as "User id not found" are ones that I don't have. I can search the keyserver either by going to a web interface such as http://pgp.mit.edu/ using gpg --search-keys tty/10 1003 erwin 10:03:16 $ gpg --search-keys 0xF74F343D gpg: searching for "0xF74F343D" from HKP server pgp.mit.edu Keys 1-4 of 4 for "0xF74F343D" (1) Richard Johnson <rjohnson(a)ucar.edu> 1024 bit RSA key F74F343D, created 1995-08-18 (2) Richard Johnson <rjohnson(a)medialab.com> 1024 bit RSA key F74F343D, created 1995-08-18 (3) Thawte Freemail Member <rjohnson(a)ucar.edu> 1024 bit RSA key F74F343D, created 1995-08-18 (4) Richard Johnson <Richard.Johnson(a)medialab.com> 1024 bit RSA key F74F343D, created 1995-08-18 Enter number(s), N)ext, or Q)uit > 1 gpg: key F74F343D: public key "Richard Johnson <rjohnson(a)ucar.edu>" imported gpg: Total number processed: 1 gpg: imported: 1 (RSA: 1) pub 1024D/E580B363 1997-11-09 Theo Van Dinter <felicity(a)kluge.net> sig F74F343D 2003-11-01 Richard Johnson <rjohnson(a)ucar.edu> sig C52DCE75 2003-11-01 [User id not found] sig 314514EE 2003-11-01 [User id not found] [etc...] If I know who to look for in between , I can then find a path to the signature I'm looking for. Unfortuatelly, there's no good way that I know of to automatically find the shortest trusted path between to keys. -- Frank Sweetser fs at wpi.edu WPI Network Engineer GPG fingerprint = 6174 1257 129E 0D21 D8D4 E8A3 8E39 29E3 E2E8 8CEC

May 21, 2004