Re: [expert] problem running sound as non-root user?
On 03-Apr-2001 David E.Fox wrote: > I'm surprised no one has mentioned the permissions on the > audio devices themselves. What, for instance, are the permissions > on /dev/dsp? If it's set to (for example) -rw-- root audio then no > wonder you can't use the device as a non-root user. > > You could circumvent the issue and just chmod 666 your audio > device files (/dev/audio, /dev/sequencer, /dev/dsp etc.) if you don't > expect someone telnetting in and sending stuff out your sound card in > the middle of the night and waking you up :). > [snip] If you didn't set them to 640 > (i.e., > -rw-rw--' then you wouldn't have been able to access them even if you > did > change the group info. > Well, we are all working with the setup mandrake gives us. I have a stock LM 7.2 install and you will note below that Mandrake made the one user (sundance) the owner of the "special" [audio] files (it might be worth remembering that /dev or device files are special files, none of which are executables per se.) crw---1 sundance audio 14, 14 Sep 27 2000 admmidi0 crw---1 sundance audio 14, 30 Sep 27 2000 admmidi1 crw---1 sundance audio 14, 46 Sep 27 2000 admmidi2 crw---1 sundance audio 14, 62 Sep 27 2000 admmidi3 crw---1 sundance audio 14, 12 Sep 27 2000 adsp0 crw---1 sundance audio 14, 28 Sep 27 2000 adsp1 crw---1 sundance audio 14, 44 Sep 27 2000 adsp2 crw---1 sundance audio 14, 60 Sep 27 2000 adsp3 crw---1 sundance audio116, 0 Sep 27 2000 aload0 crw---1 sundance audio116, 32 Sep 27 2000 aload1 crw---1 sundance audio116, 64 Sep 27 2000 aload2 crw---1 sundance audio116, 96 Sep 27 2000 aload3 crw---1 sundance audio116, 1 Sep 27 2000 aloadSEQ crw---1 sundance audio 14, 13 Sep 27 2000 amidi0 crw---1 sundance audio 14, 29 Sep 27 2000 amidi1 crw---1 sundance audio 14, 45 Sep 27 2000 amidi2 crw---1 sundance audio 14, 61 Sep 27 2000 amidi3 crw---1 sundance audio 14, 4 Sep 27 2000 audio0 crw---1 sundance audio 14, 20 Sep 27 2000 audio1 crw---1 sundance audio 14, 36 Sep 27 2000 audio2 crw---1 sundance audio 14, 52 Sep 27 2000 audio3 crw---1 sundance audio 14, 7 Sep 27 2000 audioctl The "c" indicates a character device: the whole thing is, first, a MAKEDEV issue. Only after that does it become a chmod issue. But still, the "rw" perm on the device is obviously sufficient for me (user==sundance) to have full access to sound (since I own it) and is the way Mandrake set things up. It's not quite clear whether the suggestion is that this represents a potential remote security exploit, but if so, then that is surely a mistaken suggestion. /dev files are all *special* files, and none of them are directly executable AFAIK. The thing with the above setup, all other things being equal, is that adding a new user after the fact, say sundown, to the audio group would likely be a "no joy" affair. Had I added several users during the LM install, it seems like the owner would have been root, the group "audio", all users put in the "audio" group, and the perms something like crw-rw. Does anybody else find it a bit odd the way Mandrake set this up for the lone sundance user? b5dave - 02-Apr-2001 23:16:37 -
RE: [expert] localhost address and cups conflict
On 22-Mar-2001 Jeff Malka wrote: [snip] > Anyway when I try to install a printer under kups it won't let me > saying it can detect no parallel ports. Yet in /dev there are files > named lp0, lp1 and lp2. So it is a matter of cups not seeing my > parallel port. Ditto. > Then I had a brainstorm that may be the clue to the problem based on > the error message above. > > Shortly after I installed Mandrake 7.2, I wanted to customize it and > changed the default name "localhost.localhost" (which would show up on > the top of some utility apps) to "Jeff_PC". > > Now, when I enter in a browser the URL http://localhost:631 it could > not find the cups page. BUT, when I tried http://jeff_pc:631/admin, > that opened up the admin page for cups. So, maybe cups is not working > because it's files are hardwired to look for "localhost" where now I > have "Jeff_PC". Yep, had the same problem originally. > Problem is I no longer remember how I changed the localhost name or > what the original default was (localhost or localhost.localhost). I do > recall that Drakeconf accepted the change but did not implement it and > I had to change an ascii config file somewhere. > > Anyone able to help me out of this muddle I got myself in? [snip] You're on the right track. Note that I no longer use cups for several reasons, but when I did get it working the problem was indeed a hostname conflict. Your starting point would be /etc/hosts. On a standalone machine, after the initial install, it will look like: 127.0.0.1 localhost.localdomainlocalhost the format being Yours might now look like: 127.0.0.1 localhost.localdomain Jeff_PC or perhaps: 127.0.0.1 Jeff_PC.localdomain Jeff_PC (I'm not sure just what the Mandrake tool changes) Note that all the network services read this file on boot up and modify their own config files accordingly. Personally, I suspect CUPS screws up in this regard: I had to explicitely specify the hostname in /etc/cups/cupsd.conf to get things working. This also requires restarting cups as root: # /etc/rc.d/init.d/cups restart Be careful, though, if you plan on manually changing /etc/hosts without rebooting. You would have to restart networking, and perhaps other individual networking services ... postfix comes to mind. # /etc/rc.d/init.d/network restart Although I despise rebooting, changing the hostname is one time when it can save some headaches. Also, when you run Kups, there's a menu item "configure hosts" where you can enter the hostname. Even when I got that right, I could not configure my printer via Kups ... same crap as you got: "no parallel port on this computer" nonesense. The web tool, however, did find lp0 (lpt1) no problem. I think kups is broken ... it seems to think lpt1 is lp1 even though your parallel port devices start at lp0. Hopefully the above hints are sufficient for you to get cups working. Please log everything you try, and if you succeed, please tell us exactly what you did to get it working. Thanks. b5dave. - 22-Mar-2001 22:56:09 -
RE: [expert] Lyx doesn't work with CUPS
Praedor, >If I try to print directly from lyx with the lpr command, nothing >happens. The first place to look for CUPS related gotchas & fixes is this series of FAQ's: http://www.mandrakeforum.com/mysearch.php3?author=till A search on "lyx" there narrows it down to FAQ (7). Your problem is documented and there's a fix. Peace, b5dave
RE: [expert] Problems with 2.4.0 and the reiserfs and supermount
For what it's worth, ReiserFS was only merged into the 2.4.1-pre8 kernel yesterday. (see: http://linuxtoday.com/news_story.php3?ltsn=2001-01-18-004-04-NW-KN) It never was in the 2.4.0 source. This may not help your supermount patching though. Just passing on some info; my kernel patching experience is limited to smooth sailing ... no troubleshooting. Dave. [snip] >Even though the install let me install the os on my reiserfs partitions, >the kernel that were booted from (2.4.0) didn't seem to have reiserfs in >it, since it errored on mounting my root disk (seems it tried to mount >it as ext2, even if I haven't had ext2 on this machine at all, in fact, >the only fs that has ever been on any of my new disks is reiserfs) [snip] >The reiserfs patch and the supermount patch can't be applied to the same >kernel, I can only use one of them, because when trying to add the second >I get errors (this I can understand, files are being changed by one of >the patched and the other one doesn't recognize the files).
Re: [expert] Linux worm...?
The list went down, although there seem to be conflicting reports just as to when (see the thread "Security Lists"). I received the "slocate" advisory on Dec 18, and then nothing untill today (the glibc advisory). Yet there were very many Mandrake advisories during that period that I would see posted on linuxtoday. Vincent seems to have fixed the problem (god bless him). Personally, I suspect their list software, "sympa", sucks rocks. There are newbies out there administering fairly high volume lists without a hitch, yet I'm sure our gracious Mandrake friends have a far higher level of expertise. All the Mandrake lists seem buggy as hell; there are often long delays is posts being listed (up to 1+ days), so you'll find a buch of people responding to some post, and we end up with a cascade of redundant answers. I've had at least three posts that just fell through some crack. It's got to be the sympa software. >I did subscribe to the Mandrake security announcement list, but I never >get anything from it. Whats up with that?
RE: [expert] Linux worm...?
Yes, I know this list isn't the forum within which to debate this issue, but there is a great danger in our midst, and I promise to be terse. Marc, what you desire and suggest corresponds exactly with the mindset of those powerful entities who want to establish ubiquitous e-commerce on *our* internet: they want to own and control the internet. They represent *big* money and influence *big* government. The collection of "evil" crackers, script kiddies, and spammers is *nothing* compared to the entities that are lobying to rule *our* internet. Those large commercial entities will invoke the "evil" of the crackers and spammers for their purposes only, and you can bet that their proclamations will be cleverly veiled in rhetoric about "maintaining the purity of the Internet". I think we should focus on security. Internet-wide systems and controls designed to hinder crackers are just what big business wants. I think we will be seriousy fucked if big business manages to implement such systems and controls. And if this Armageddon comes to be, I'll be pointing the finger at the multitude of Linux distros that even allowed insecure setups, and at the moment, that's just about all Linux distros. b5dave On 18-Jan-2001 Mark Weaver wrote [snip] > Seriously though...it's about darn time that something SERIOUS be done > about and WITH these people that are a great big pain in the arse to the > rest of the world that HAVE a real life and a descent direction for that > life. [snip]
RE: [expert] ram issue(continued)
chronos, I'll tell you right off the bat that I'm in a real foul mood tonight, and I'm not at all certain why. But responding to your post might be theraputic (sp??). #1: The ability to edit files is essential in any Unix variant, be it Linux, Solaris, BSD, etc, etc. #2: The Operating systems mentioned or suggested in #1 are only suitable for those people who have the willingness and ability to learn how to edit files (at the very least!) #3: Most of the people in the world do not have either the willingness nor the ability to learn how to edit a file. #4: Your posts suggest that you are like most of the people in the world. #5: Most of the people in the world would find Windows or the Macintosh more suitable. #6: As #5 will have many people on this list in fits, just let me say that for every computer literate type, there are two associated parents, four associated grandparents, innumerable sisters and brothers, cousins, nieces and nephews, etc. To suggest that tech savy, or even semi-competent computer users are in the minority should be both an obvious and trivial observation. #7: Given #1 - #6, and if they obtain in your case, I think it's reasonable to ask why *you* are choosing the Linux route? I know my mother is far from being expert "pointer-and-clicker", yet even the expert "pointers-and-clickers" from the Windows world, as certified and highly paid as they sometimes are, are completely lost when computer use means brain use. #8: In Linux, computer use means brain use. In Windows, that is not necessarily the case. If you want your computer to be nice & easy, then , by all means, stick with Windows. That is what Windows was meant for. The same goes for the Mac. Just because Linux users need to be be at least semi-smart is no reason why you should be denied a satisfying PC experience. Microsoft and Apple design computers for most of the people in the world, people just like you. Peace b5dave. On 19-Jan-2001 chronos . wrote: > To Chris- > No I have absolutly no experience editing files. So go flame away. I am > a newbie and I dont understand sue me. > You started somewhere too. Kick me off the list for all I care. Excuse > me for not being a genius. > On Thu, 18 Jan 2001 15:00:05 -0500 "D. Stark - eSN" > <[EMAIL PROTECTED]> wrote: >>Except I think the fella is using grub. >> >>Chronos: >> >>First off, run this from a command line: >> >>[dstart@fweeble ~]$ cat /etc/lilo.conf >> >>Just the cat part, mind you. If you DON'T have a file called >>/etc/lilo.conf, >>you installed grub as your boot loader. All of the great advice >>everybody >>has offered up to this point is for naught. Were I you, at this point, >>I'd >>say screw it all and reinstall, this time choosing LILO as your boot >>loader >>(why did Mandrake include grub? freedom of choice?). Were I you. Who >>knows, >>it may solve the ram issue. If it doesn't then Chris' great advice >>below >>will work. >> >>You're a newbie, I betcha. Don't worry, we all start somewhere. I was >>lucky >>enough to room with a net.god in college. I know sending people to >>LinuxDoc.org is my answer to most tech questions, but you might want to >>spend some time reading up here: >> >>http://linuxdoc.org/LDP/gs/ >> >>also: >> >>http://63.209.80.231/en/doc/72/en/ref.html/index.html (mandrake users >>manual) >> >>Seriously, a good place to go to keep from killing yourself. If you >>would >>like a good deadtree guide, check out O'Rielly's >>_Learning_the_UNIX_Operating_System_. It's a very good book to get you >>up to >>speed on the basics. From there the specifics become a lot easier to >>handle. >>Make sure to learn to use 'grep' and 'less' as well. They're your >>friends. >> >>Derek Stark >>IT / Linux Admin >>eSupportNow >>xt 8952 >> >>-Original Message- >>From: [EMAIL PROTECTED] >>[mailto:[EMAIL PROTECTED]]On Behalf Of Chris Spencer >>Sent: Thursday, January 18, 2001 2:29 PM >>To: [EMAIL PROTECTED] >>Subject: Re: [expert] ram issue(continued) >> >> >>On 18 Jan 2001, chronos . wrote: >> >>> Hi all, >>> I`m running mandrake 7.1 and I have 128 for ram but it only sees 64 >>> of my >>128. I went to mandrakeuser.org and found the following- edit tittle >>linux >>kernel (hd0,0)/boot/vmlinuz root=dev/hda1 then append mem=128M to the >>end of >>it. The problem is I type pico then tittle linux and it gives me a new >>file >>then I type kernel etc. and all it does is places a file
Re: [expert] Cups is winning the battle
Jerry, It would appear I'm totally irritated and irritating tonight (sigh). Glad you got it sorted out, and replying to your own e-mail is not bad form if it adds useful and relevant info to your original post, which, in this case it does. But the form of your post is rather incoherent, as you are attributing to me, at least, the entirety of *your* last post. Maybe it's the way the quoting in your email program is set up. But, hell, I can live with that. I really wanted to know if my suggestions regarding your /etc/cups/cupsd.conf file were of any value. On 19-Jan-2001 Jerry Sternesky wrote: > On Thursday 18 January 2001 20:39, Jerry Sternesky wrote: > OK Bad form replying to your own email. [and Jerry's previous post] > On Tuesday 16 January 2001 10:41, b5dave wrote: > OK cups is alive and staying alive. I also get the ghostscript rpms > t-shirt, I forgot I had the install cd in the drive and after select > printerdrake. Bam! I was a member of the club. [etc snip] No, b5dave (me) didn't write any of that! I'm not dissin' you, Jerry, and I can appreciate the euphoria of overcomming CUPS problems, but your quoting of our exchange has even *me* confused as to who said what. I like to think of these exchanges as reference material for other users. The issue of whether one should place the quote before or after the reply may never be resolved, but both methods can be clear about who said what. peace ... love LSD. It's all good. Sorry about my complaining. It's all good. (I guess I should now sign all my posts as b5dave since there are several other "Dave's" on the list ... and perhaps update my medication ...) peace, b5dave
[expert] To mail list experts re this list
Looks like I'm complaining again (sigh) about a free, perhaps volunteer service from which I have no real right to expect anything. Still... I'd really like to know if it's just me, or are we all in some significant time warp here. I've had posts on this list delayed more than 24 hours, and a few haven't shown up at all. Today, I've been posting since 17:32 pm and it's now 20:25 pm; none have shown up yet. I've never experienced a 3 hour delay on any other list; 5-7 minutes was the longest delay I've known, and 30 secs to 1 minute has been the norm. Sure, instant gratification or resolution is desirable in our times, but there are other consequences that these delays entail. Consider my post of Jan 17th with subject "Failed message delivery". O.L. Plaine Jr provided the explanation. For many hours after this, others responded that they were experiencing these same "Failed message delivery" messages. Then Chris Spencer essentially repeated Mr. Plaine's explanation. The result of the approx 3 hour+ delay on the list resulted in 3 redundant messages out of a total of 5 messages. Aside from my initial post, nobody knew what anybody else was posting during "the time warp". Now, anybody who follows this list, and who takes a moment to think about it, will realize that every single thread suffers the same sort of redundancy. I have no math genius, but I think that the volume of traffic on this this could be less than half of what it is if the typical delay were reduced to some reasonable level. Moreover, the reduced redundancy would make it a much more valuable resource and reference. So much for the philosophy. My pragmatic side tells me something's askew here, and, assuming that the Mandrake people who maintain these lists care about what they do (I have every reason to believe they do very much), then I'm compelled to think that the list software (sympa) is very defective (a 3 hour delay is surely not bandwidth related, right?). My question, then, is just what you "mail list software" experts out there think about all this? Is "sympa" broken? Should we petition Mandrake to change list software? Am I being completely unresonable? Has Rod Serling moved into my appartment? Clearly this delay business is driving me nuts, and I really need some sympathy. BTW, "sympa" is French slang or short for "sympathique", meaning friendly, or nice; it's somewhat different from the English "sympathetic". peace, dave.
Re: [expert] Security Lists
Okay, well the Security Announce is working for me gain; just got the glibc advisory. Thanks for the work Vincent, and sorry if I was overly critical. Dave. Vincent wrote: > At this point the going is slow to find a fix because that individual > is gone for the week, but rest assured we are trying to get this fixed > ASAP.
Re: [expert] RPM updates
-Uvh will update rpms that already exist on your system, but if the rpm doesn't already exist on your system, it will do the same as an -ivh (i.e., install the new rpm). In this sense, -ivh is redundant. -Fvh will *only* update rpms that already exist on your system. Dave. . On 18-Jan-2001 Mark Weaver wrote: > So then it's better to do the -Uvh instead of the -Fvh?
[expert] Failed message delivery
Hi folks, Since a few days, whenever I post to the list I get an e-mail from [EMAIL PROTECTED] that starts like: The msg to [EMAIL PROTECTED], had delivery problems due to unknown recipient or host: [etc etc etc] This is too weird. My posts work, but who the hell is [EMAIL PROTECTED] and why are my posts being bounced to that address? This only happens on this list; all other e-mails or other lists work fine. Anyone else getting this? -- Thanks, Dave.
RE: [expert] Win4Lin 2.0 - J'ai 2 répertoires /usr/src/linux/
Tout est beau; c'est normal. Le répertoire linux est un lien symbolique à linux-2.2.17. Le système va toujours chercher pour le répertoire linux. Ce dernier pourait, par exemple, etre lié à une différente version du "kernel" pour raisons d'expérimentation. Si tu fais un "ls -la" dans le répertoire /usr/src tu vas voir quelque chose comme: $ ls -la total 3 drwxr-xr-x5 root root 128 Dec 11 05:46 ./ drwxr-xr-x 20 root root 440 Dec 11 12:59 ../ drwxr-xr-x7 root root 141 Dec 11 05:19 RPM/ lrwxrwxrwx1 root root 12 Dec 11 05:36 linux -> linux-2.2.17/ drwxr-xr-x 17 root root 575 Dec 11 05:36 linux-2.2.17/ drwxr-xr-x4 root root 81 Dec 11 05:46 php-devel/ Dave. [] /usr/src/linux-2.2.17 contenant : include/ README.kernel-sources et /usr/src/linux (en italique d'ailleurs... je ne sais pas ce que cela signifie) contenant également : include/ README.kernel-sources Résultat, je ne sais pas lequel utiliser, s'il y en a un de trop, si tout y est, bref, je suis paumé. Quelqu'un pour guider l'aveugle ? Merci.
Re: [expert] Security Lists
Vincent, > No, that is not the case at all. It's not useless and it's not > dangerous. We're just having some difficulty with it. Sorry, but I must disagree. One expects to be the first notified of Mandrake security issues when one is subscribed to Mandrake's security-announce. There's an implied if not explicit expectancy that the list should be sufficient for being alerted to Mandrake security issues. I'm not saying the list has always been useless and dangerous, and I'm not saying the list will always be such. But the list has been down for just about a month, and as far as I can tell, is still down. So at present, assuming the list is still down, it is indeed useless and dangerous simply because people expect it to be both functional and current. It is analogous to some virus or trojan disabling my system's local security warnings. peace, dave.
RE: [expert] Removing SCSI devices?
Nice! It's all good. Thanks. The reason html mail is problematic has to do with one of the basic *nix philosophies, i.e., one tool for one job. Every *nix/linux mail program expects (more or less) to get some sort of text format. Html, on the other hand, is generally handled by a web browser. Consequently, if one is using a real and bonafide mail program, html mail will end up looking something like this: - > > My system(Mdk7.2) was installed with SCSI option. > However, > >since I > >decided to remove the SCSI drive, I just removed the > drive(while > SCSI > >adapter is still plugged in) then immediately I get the > kernel > panic at > >boot. > > > > What do I have to do for the system to boot up when the > SCSI > drive( and > >adapter) is removed? > > > > Thanks > > > > > > -- All this nonesense, I'm sure, is the result of Microsoft shenanigans. If you are posting from Outlook*, then it wouldn't be surprising that M$ has the web browser integrated into it, and everywhere else it can: that, indeed, was the strategy M$ used to wipe out Netscape, and the basis of the DOJ's/States' legal action. Note also that M$, and even Netscape, often default to sending a plain text email message with an html copy as an attachment. This too is a *bad* thing. Turning *off* the html attachment is a *good* thing. Plus you end up using way more badwidth than is needed, and this affects us all: wasting or abusing bandwidth is a bad thing. Peace, dave.
Re: [expert] Security Lists
if it's any help, the last advisory I got was the "slocate" one of Dec 18/2000. dave. > We're looking into it. I have my suspicions that something has > changed with sympa and it is rejecting the mails silently so this > didn't actually come to my attention until about two days ago. We > hope to have it fixed ASAP.
RE: [expert] Security Lists
Matthew, > Has anyone else been seeing traffic on the Security lists? I joined both the Mandrake security announce and security discuss lists just before the new year, and there was some brief traffic. Since then, however, nothing. Last week linuxtoday (http://www.linuxtoday.com/) was full of Mandrake security updates, and there has yet to be a peep about these on the Mandrake security lists. It seems that these lists are pretty useless, if not dangerous. dave.
RE: [expert] Cups is winning the battle
On 16-Jan-2001 Jerry Sternesky wrote: > It seems I am having lots of problems with cups. When I installed > Mandrake 7.2, it found my epson 777 printer and set it up as a 740. > There were no drivers for 777. I got a great test page and since then > I haven't been able to do a thing with it. My LM7.2 install didn't find my old HP Deskjet 520 on lp0 so I tried to setup printing post install. I'm on a simple single non-networked box. > It seems that the cups dameon just stops running. I was getting the same thing. IIRC, somewhere in the docs it mentioned that some minutes after you start the daemon, it checks back with /etc/cups/cupsd.conf but I could be way off base here. Nonetheless, manually editing cupsd.conf fixed the dying daemon problem for me. None of the config tools would worked because the install left me with an unusable cupsd.conf file entries like ServerName were still remmed out. I could start the daemon (# /etc/rc.d/init.d/cups start), but, like you say, it would die after a few minutes. > After I start the machine I do a service cups status, I get back it's > running and a pid number. After about 5 minutes, the status returns > cups is stopped. I restart it and 5 minutes later it's stopped. I am > really at a loss here. Unless Till (he's the resident CUPS guru) has the fix for you, I'd suggest you first take a quick look at your /etc/cups/cupsd.conf file ... along with a 'man cupsd.conf'. > When I use a client and netscpae to connect to http://"server":631 I get > to the cups setup and the printer is there and configured, but I am > unable to print a test page because server.domain.com not located, so I > know I have a network issue to resolve with addressing I could well be wrong, but this sounds more and more like a bum entry in your cupsd.conf file. > Since I will be scouring how-to's and faq's if anyone has some good > ones besides the ones on mandrakeuser.org to point me to I would > appreciate it. There's a pile of FAQ's here: http://www.mandrakeforum.com/mysearch.php3?author=till plus the various resources at: http://www.cups.org > BTW this machine is current as of 1/15 with all the updates for 7.2. Pretty well the same thing here, so . I should also mention that I'm pretty certain PrinterDrake or Drakprint or whatever it's called is broken. Going that route, you'd be asked to insert your install CD, and, without warning, the program would proceed to wipe out your updated Ghostscript rpms, replace them with the ones from the CD, and promptly barf. Myself and at least two others have "been there, done that". Do let us know how you get this worked out because, as you can see, the whole mess isn't particularly clear to me either, and there aren't a whole lot of people jumping in with concise solutions. Personally, I'm ready to revert back to the old lpd/lpr system because the two fonts I most often use for printing from Wordperfect come out slightly better the old way. -- dave.
RE: [expert] Removing SCSI devices?
Okay, I'll be the bad guy. Yo, Kim TK, sending html mail to a list is considered by many to be rude, offensive, a bad idea, in bad taste, and may also be a loss of face on your part. You will have a better chance of getting help by not offending the people you are seeking help from. I'd suggest you try to figure out how to turn off your html mail; the world would be a better place. :-) Peace, dave.
Re: [expert] PLEASE FIX REPLY-TO in this list!!!
Bob is talking specifically about the *digest* version of the list. My "reply-to's" work fine, but I *don't* get the digest version. Is there anyone here with the digest version with the same problem as Bob? I may be wrong, but I'm assuming that Viktor didn't catch the *digest* aspect of Bob's complaint. If I'm wrong, Viktor, please accept my apologies. On 15-Jan-2001 Viktor Lakics wrote: > This is a *REPLY* to your message. It goes to > [EMAIL PROTECTED] I think your email client does not > handles that correctly...FYI, I use mutt under LM7.2. > > Viktor > On Mon, Jan 15, 2001 at 03:28:02PM -0500, Bob Puff@NLE wrote: >> Would someone PLEASE fix the REPLY-TO address for the digest of this >> list?! Replies get sent to "expert-request", which then fail. >> >> Bob >>
[expert] Cups removal precedure?
Okay, look, I don't want to get into the plusses or minusses of CUPS. I've paid my CUPS dues, and would like to go back to lpr for now. Sure I think lpr is brain dead, and I think CUPS is the future of printing in *nix. But for *now*, I would really like to wipe CUPS from my system and replace it with the lpr system. I have a stock LM 7.2 install (with the latest security upgrades). Can someone please tell me what CUPS packages I need to "rpm -e" and what lpr packages I need to "rpm -Uvh --test" ?? Please? Thanks. Dave.
Re: [expert] Adding TrueType fonts to LM 7.2
Hate to be a party pooper here, but adding fonts to LM 7.2 and adding fonts to Wordperfect (7-8) for Linux are two entirely different things. In Wordperfect, font rasterization (both display and printing) is handled by Wordperfect exclusively, and it does not support True Type. But you can find Type 1 fonts with better 'hinting' than those supplied on the Wordperfect for Linux CD; indeed the Type 1 fonts supplied with Wordperfect for Windows, with OS/2, and with Applix are generally superior. There is, however, at least one utility to convert tt fonts to type 1 if you're feeling adventurous. The best documentation I've found on the subject is here: http://members.bellatlantic.net/~smithrod/wpfonts.html For other WP for Linux docs, see: http://www.qwkscreen.com/WPLinuxLinks.html Dave. On 10-Jan-2001 Marsden MacRae wrote: > "D. Stark - eSN" wrote: >> >> http://linux.com/tuneup/database.phtml/X11/002036.html >> >> Derek Stark >> IT / Linux Admin >> eSupportNow >> xt 8952 > > Thanks! that was very helpful. We use Linux WordPerfect here, and the > lawyers want to add fonts. > > Marsden
Re: [expert] problems with network
As root try: # services --status-all Any Chain input will be listed. And if you see a line that says "USAGE: pmfirewall [command]" then your uninstall didn't work. If you need to manualy uninstall, you'll find it under /usr/local/pmfirewall, as well as references in your rc scripts: #grep -i -r pmfirewall /etc/rc* I use it with a ppp connection so it is started in /etc/ppp/ip-up.local. With eth0 I figure it would just be started in /etc/rc*, but I could be wrong about that. If it is indeed running, it must be misconfigured. I remember there being options for a gateway, as well a specific IP's to allow. I don't have a network, so I can't help you there.I've had great success with the program however. Good luck. Dave. On 02-Jan-2001 Tib wrote: > I stopped it, uninstalled it, and flushed all the ipchains rules. It > /should/ > be a clean setup again like I had before. But still it acts like I said > before: only icmp gets forwarded through it and data in general is > extremely > slow. > > > Tib > > > On Wed, 3 Jan 2001, Mike MacCana wrote: > >> You can stop pmfirewall when its installed by running >> >> service pmfirewall stop >> >> Which clears the rules and set them to deny. Did you read the >> documentation fully? Its a neat utility, so don't give up on it yet. >> >> If you must use ipchains directly, then... >> >> ipchains -F input >> ipchains -F output >> ipchains -F forward >> >> Wil do the trick... >> >> Mike >> >> -- >> Mike MacCana Support Consultant >> C Y B E R S O U R C E >>Level 9, 140 Queen St Melbourne 3000 >> Ph : +61 3 9642 5997 Fax: +61 3 9642 5998 >> >> On Tue, 2 Jan 2001, Tib wrote: >> >> > Ok! here's some fun that's been causing me to lose my hair in chunks >> > because >> > I've been ripping it out. I checked out PMFirewall as someone >> > mentioned, and >> > had initially set it up. Everything seemed fine. However, even though >> > I ran >> > ipchains -L and it said there were no rules in place, it seems that >> > NOTHING but >> > icmp packets are allowed through the network anymore. I've smashed my >> > head >> > against my route table and other things but I just can't figure it >> > out. Nothing >> > gets forwarded but icmp, everything else is no route to host (from >> > the machines >> > inside the network that would use the address for eth0 as a gateway), >> > furthermore, it would seem that eth0's performance has been >> > compromised as well >> > for some strange reason, data throughput of ANY kind is extremely >> > slow even >> > though pingtimes are fantastic. I uninstalled pmfirewall and backed >> > out every >> > other change I can think of, I've shut down and brought up the >> > network, >> > re-added routes the way they were before, and overall exhausted my >> > brain with >> > this problem, can someone help please??? >> > >> > >> > Tib >> > >> > >> > >> > >>
RE: [expert] KMail: import files (OT?)
What you've got with your xfmail setup, as you've described it, are "mh" type mailboxes. I have the same thing. You want to convert your "mh" mailboxes to "mbox" type mailboxes. The docs for Kmail talk about a conversion utility (mh2kmail) "that's included with the source", which seems to mean cvs, as Kmail is otherwise only available as part of the kdenetwork package. For other means of conversion see: http://kmail.kde.org/manual/secimporting.html (not much help) and man packf (this seems to be *the* utility, but I've not used it) dave . On 29-Dec-2000 Guido Milanese wrote: > I do not know if this question is OT -- if so, please forgive me. > I had been using xfmail for some time, after migrating to Linux. Now I > find the new version of KMail very stable and reliable, so I'm using > this other program. > > The problem: xfmail archives mail making a folder for each category you > want, and mail is stored with numbered files under each directory (e.g. > /home/guido/mail/mandrake/1, /home/guido/mail/mandrake/2 and so on). > > Kmail does not use one file for each message, but archives mail using > one "large" file for each category, so there will be only one > directory with "large" files, e.g. /home/guido/Mail/mandrake. > > Question: is there a program to merge the old mail files (xfmail > output) with the new ones (kmail output)? or should I write a little > utility do perform this task? > > thanks! > guido, italy > > -- > E-Mail: Guido Milanese <[EMAIL PROTECTED]> > Vocal Ensemble Ars Antiqua, Genova, Italia > Homepage: http://fly.to/arsantiqua > + + + + + + NON NOBIS DOMINE + + + + + + + > --
Re: [expert] Recursive symlinks in /usr/bin?
Weird. I *do* have /usr/bin/mh which seems to duplicate the contents of /usr/bin, and both directories have mh -> ./ . A security thing maybe?? Dave. On 29-Dec-2000 Larry Marshall wrote: > Tom Eastman wrote: >> >> [tom@celleste bin]$ ls -ld /usr/bin/mh >> lrwxrwxrwx1 root root1 Dec 29 2000 /usr/bin/mh -> >> ./ >> >> Why is there a symlink in /usr/bin called mh which points straight back >> to /usr/bin??? >> >> Is this kind of recursive symlink a bad thing? or can I just ignore >> it? > > Can't imagine why it's there or how it could be useful. For what it's > worth, I have no such link on my system. Any idea what mh is or is > supposed to be? > > Cheers --- Larry
Re: [expert] Desktop device icons, folders, etc, don't work
Wow! Is that the result of installing every language available? On 24-Dec-2000 Pierre Fortin wrote: > Praedor Tempus wrote: >> >> Could someone please use an editor and view the device kdelnk in their >> Desktop directory (preferably from KDE 2.0, 2.0.1 or 2.1) and post the >> contents? > > Without a CD in the drive, I get the non-intuitive: > Error -KDesktop > Unable to enter file:/mnt/cdrom. > (X) You do not have access rights to > this location. > [ OK ] > > With a CD, it's fine. > > KDE2.0... > > Cd-Rom.kdelnk: ># KDE Config File > [KDE Desktop Entry] > Comment=CD-ROM device > Comment[br]=Trobarzhell lenn CD-ROM > Comment[ca]=Dispositiu de CD-ROM > Comment[cs]=CD-ROM mechanika > Comment[da]=CD-ROM Enhed > Comment[de]=CD-ROM Laufwerk > Comment[es]=Dispositivo lector de CDROMs > Comment[et]=CD-ROM seade > Comment[fi]=Tiedostosysteemi > Comment[fr]=Périphérique lecteur de CDROMs > Comment[ga]=Gaireas Dlúthdhiosca ROM > Comment[gl]=Dispositivo CD-ROM > Comment[hr]=CD-ROM ureðaj > Comment[hu]=CD-ROM meghajtó > Comment[id]=CD-ROM Anda > Comment[is]=Geisladrif > Comment[it]=Lettore CDROM > Comment[lt]=CDROM árenginys > Comment[nl]=CDROM drive > Comment[no]=CDROM enhet > Comment[pl]=Urz±dzenie > Comment[pt]=Dispositivo com Sistema de Ficheiros > Comment[ro]=Lector CDROM > Comment[sk]=CD-ROM mechanika > Comment[uk]=ëÏÍÐÁËÔ ÄÉÓË > Comment[wa]=Éndjin lijheu di plakes lazer > Icon=cdrom_supermount.xpm > Type=Link > URL=/mnt/cdrom > > Floppy.kdelnk: ># KDE Config File > [KDE Desktop Entry] > Comment=Floppy Device > Comment[br]=Trobarzhell pladennigoù > Comment[ca]=Dispositiu de disquet > Comment[cs]=Disketová mechanika > Comment[de]=3½-Zoll Laufwerk > Comment[es]=Dispositivo disquetera > Comment[et]=Flopiseade > Comment[fi]=Tiedostosysteemi > Comment[fr]=Lecteur de disquettes > Comment[ga]=Gaireas flapach > Comment[gl]=Dispositivo lector de disquetes > Comment[hr]=Disketni ureðaj > Comment[hu]=Hajlékonylemez meghajtó (floppy) > Comment[id]=Isi Disket Anda > Comment[is]=Disklinga drif > Comment[it]=Unità floppy > Comment[lt]=Diskeliø árenginys > Comment[nl]=Floppy drive > Comment[no]=Diskett enhet > Comment[pl]=Urz±dzenie > Comment[pt]=Dispositivo com Sistema de Ficheiros > Comment[ro]=Lector de dischete > Comment[sk]=Disketová mechanika > Comment[uk]=äÉÓËÅÔÁ > Comment[wa]=Mineu di plaketes > Icon=3floppy_supermount.xpm > Type=Link > URL=/mnt/floppy > > HTH, > Pierre - 24-Dec-2000 10:39:38 -
RE: [expert] Desktop device icons, folders, etc, don't work
On 24-Dec-2000 Praedor Tempus wrote: [snip] > Could someone please use an editor and view the device kdelnk in their > Desktop directory (preferably from KDE 2.0, 2.0.1 or 2.1) and post the > contents? Specifically, if you have a functional floppy, zip, or > harddrive device icon on your desktop, please post the text contained in > the kdelnk for in it Desktop/. > > For my non-functional Zip Drive kdelnk, for instance, it reads: > > [Desktop Entry] > Dev=/mnt/zip > FSType=supermount > Icon=3floppy_mount > MountPoint=/mnt/zip > Type=FSDevice > UnmountIcon=3floppy_unmount [snip] Sure. Note that I don't use the KDE environment, but I did install the whole thing so that I could use all the KDE apps from the Enlightenment WM. So the kde config files I'm posting are *unaltered* from the initial install of 7.2 and may be of some use to you. The following are under ~/Desktop. $ less CD-ROM [Desktop Entry] Icon=cdrom_unmount Type=Link URL=/mnt/cdrom $ less Floppy.kdelnk [Desktop Entry] Icon=3floppy_unmount Type=Link URL=/mnt/floppy peace, dave.
RE: [expert] fstab entry for supermount
On 24-Dec-2000 Praedor Tempus wrote: > In trying to fix my problem with desktop devices not working, I altered > my fstab for cdrom, zip, and floppy to eliminate the supermount and made > them the old-fashioned manual mount devices. I would like to go back to > supermount but cannot recall the syntax and there is no backup fstab > with the supermount entries in them. If it's any help, here's the unaltered supermount fstab entries from last week's 7.2 install: /mnt/cdrom /mnt/cdrom supermount fs=iso9660,dev=/dev/cdrom 0 0 /mnt/floppy /mnt/floppy supermount fs=vfat,dev=/dev/fd0 0 0 peace dave - 23-Dec-2000 22:02:16 -
RE: [expert] Unmount supermount without reboot
See one of the most explicit and simple man pages ever written: $ man supermount On 22-Dec-2000 Viktor Lakics wrote: > Does anyone know how to unmount (suspend supermounting) a supermounted > removable disk without editing out the corresponding entry in fstab and > reboot? Eg. to use ext2fsck on a zip disk? > > Thanks: Viktor > > Viktor Lakics > > Through the Internet: > > [EMAIL PROTECTED] > > > > Get free email and a permanent address at http://www.amexmail.com/?A=1 - 22-Dec-2000 04:20:46 -
Re: [expert] 'Use hard drive optimizations' does what?
On 22-Dec-2000 Andrew George wrote: > On Fri, 22 Dec 2000 13:57, A V Flinsch wrote: >> On Thursday 21 December 2000 21:17, b5dave wrote: >> > So the burning question is: Does anybody have the hdparm command in >> > their rc scripts or elsewhere??? >> >> I have the following at the end of rc.local >> >> # hdparm stuff >> echo -n "Harddrive optimizations" >> /sbin/hdparm -c 1 -d 1 -k 1 /dev/hda >> /sbin/hdparm -c 1 -d 1 -k 1 /dev/hdb >> /sbin/hdparm -c 1 -k 1 /dev/hde > > Oddly enough I discovered that even though I had a hdparm line in rc2.d, > rc3.d and rc4.d...I had to add the line to rc.local to get it to work Indeed! Hmmm. What service scripts in rc2,3,4? Runlevels 0,1,2,and 6 mean your machine is on its way down or just plain down. Runlevel 4 isn't really used. The only two runlevels of interest are 3 (which boots you into a console), and runlevel 5 which boots into the xdm graphical logon screen. In other words, rc2 and rc4 don't count. If your setup was booting into a graphical logon then you were booting into rc5, and your hdparm entry in rc2, 3, or 4 would never work. Note that the runlevels are *not* sequencial; booting into rc.5 (runlevel 5 == graphical logon) will kill all prior stuff and just start the rc.5 stuff. The files in those rc.* directories either start with an S or a K. The K means kill and the S means start. OOF, that was a mouthfull. But, again, I'm now really curious to know what scripts Mandrake was screwing with in rc.2, rc.3, and rc.4. WTF was Mandrake doing here? Dave. - 22-Dec-2000 02:34:07 -
Re: [expert] X 4.0.2 and the RPM database
Not sure what you mean by "...what looks like ..". Rebuilding the database restores/updates the dependancy issues, *tries* to resolve duplicate entries, and so forth. Are you having problems installing or "-e" ing some rpms? When you say "looks" are you being literal? Please don't think I'm being facetious here; I'm *not*. I'm just trying to cover the bases. If you want it to look ordered is some way, and right after an install an 'rpm -qa' is indeed pretty well ordered, then there are tools for that. I would certainly agree that an un-ordered "rpm -qa" is next to useless. Personally I don't find uppercase files being sorted before lowercase ones usefull at all. Plus, why call it 'Eterm' when 'eterm' does the same work? Hell, I still don't know why they called it .Xdefaults instead of .xdefaults. There, that feels better. :-/ I pretty well have a *term constantly displaying all my installed rpm's. I've used the same command since RH 4.2 $ rpm -qa | sort -fd | less It's probably too late for me to put an alias in ~/.bashrc (like alias rpms='rpm -qa | sort -fd | less') because I can type the longer version as fast as the the alias!! Anyway, if I've completely misunderstood you, I apologize: it wouldn't be the first time I've missed the mark. Yet surely it could be useful given the peculiarities of Drakupdate or whatever that dangerous contraption is called. Dave. On 22-Dec-2000 MichaelM wrote: > b5dave wrote: > >> # rpm --rebuilddb >> Don't worry, it takes a while. > > Nope. Still getting what looks like an old RedHat database =( - 22-Dec-2000 01:38:09 -
RE: [expert] X 4.0.2 and the RPM database
# rpm --rebuilddb Don't worry, it takes a while. Dave. On 22-Dec-2000 MichaelM wrote: > I'm sure I'm not the only one to be bitten by this, but for those out of > the loop, the binaries for XFree86 4.0.2 over-write the RPM database, > making RPM related stuff nigh on impossible. > > The question is, is a fix available, or even possible?
Re: [expert] 'Use hard drive optimizations' does what?
Alex wrote > I have the following at the end of rc.local > ># hdparm stuff > echo -n "Harddrive optimizations" > /sbin/hdparm -c 1 -d 1 -k 1 /dev/hda > /sbin/hdparm -c 1 -d 1 -k 1 /dev/hdb > /sbin/hdparm -c 1 -k 1 /dev/hde and . Bill wrote: >I've got mine as the last line of rc.sysinit. It works. Well if either one of you guys says that the install did indeed append that to either script, then that makes my initial question about a possible kernel parameter moot, and that's one more mystery solved. I'm perfectly happy with that. :-) ps Bill. Is that x-vcard attachment a *nix thing? I think I may have got one recently from an associate and flamed them for sending me yet another proprietary M$ attachment. OOPS! - 21-Dec-2000 22:30:11 -
Re: [expert] 'Use hard drive optimizations' does what?
Yes, I understand. In lieu of "engaged" I should have said "setup" , because somewhere there would have to be a line like: 'hdparm -m 8 -d 1 -u 1 -c 1 /dev/hda' to ensure that it is set on subsequent re-boots. If you think you have it set, just try # hdparm -tT /dev/whatever and look at the buffered disk reads. I went from something like 4 MB/sec to 19.88 MB/sec. I figure its some DMA setting in the kernel config that relates to the mobo's ide chipset. There's a whole slew of block device settings in .config which is kinda why I wanted to know. I'm positive I checked that off during the install, but hell, who knows. As Eric suggested, the only reasonable place to put the command would be somewhere under /etc/rc.d like runlevel 3 or 5 or rc.local (which is where I've put it). So the burning question is: Does anybody have the hdparm command in their rc scripts or elsewhere??? Cheers, Dave. On 22-Dec-2000 Amit Bapat wrote: > hdparam is not a daemon that runs, it just sets some parameters to be > used > for your HD > > > Original Message Follows > From: b5dave <[EMAIL PROTECTED]> > Reply-To: [EMAIL PROTECTED] > To: [EMAIL PROTECTED] > Subject: [expert] 'Use hard drive optimizations' does what? > Date: Thu, 21 Dec 2000 18:32:28 -0500 (EST) > > I initially thought that install option would engage hdparm since the > warning is the same, but hdparm was definitely not running.. That leaves > the kernel. Does anyone know what config options get set? > > Thanks > Dave. > > - > 21-Dec-2000 > 18:32:28 > - > > > _ > Get your FREE download of MSN Explorer at http://explorer.msn.com - 21-Dec-2000 21:17:22 -
RE: [expert]
The background is called by xsetroot in /etc/X11/xdm/Xsetup_0 (the comments in there suggest there are "issues" with kdmdesktop/kdedesktop.) The rest of the screen is defined in /etc/X11/xdm/Xresources Dave. On 21-Dec-2000 a r wrote: > Hi! > > Does anyone know which file contains the details of the screen you see > when > the login manager loads? > > I want to change the background and there seems to be a bug in kcontrol > that stops you from saving changes to the background eg if you want to > replace the blue screen with a .png or .jpg, (you can enter them but it > resets once you leave the login manager option). > > Thanks > > > __ > FREE Personalized Email at Mail.com > Sign up at http://www.mail.com/?sr=signup - 21-Dec-2000 20:23:51 -
RE: [expert] cups not working properly
http://www.mandrakesoft.com/~tkamppeter/packages/ Read the text file. These are LM7.2 rpms are are supposed to work 'out of the box". On 21-Dec-2000 Vic wrote: > Cool,. where can I find the url of this site? > > On Wed, 20 Dec 2000, b5dave wrote: > Also Till setup a site with upgraded >> drivers which won't be part of Mandrake updates as they are not >> security >> related. I've had no need to try these however. >> >> Dave >> >> - 21-Dec-2000 19:01:21 -
[expert] 'Use hard drive optimizations' does what?
I initially thought that install option would engage hdparm since the warning is the same, but hdparm was definitely not running.. That leaves the kernel. Does anyone know what config options get set? Thanks Dave. - 21-Dec-2000 18:32:28 -
Re: [expert] cups not working properly
The latest CUPS info by Till, along with discussions is here: http://www.mandrakeforum.com/mysearch.php3?author=till Check out FAQ(7) where there's info about the many KDE printing issues. The A4 thing seems to be a known bug. Note also that there's been a lot of fixes to all the associated CUPS files that will never be in Mandrake updates because they are not security related. The drivers and the ghostscript stuff is all very recent. Till has the stuff here: http://www.mandrakesoft.com/~tkamppeter/packages/ Although I have a crappy HP deskjet 520 (circa 1993), I tried the GIMP-CUPS driver and it looked like I was out of ink. I had fantastic results with the Foomatic drivers. For your Epson Color Stylus 600 you should try scrolling way down past "Epson Color .." to "Stylus 600" where you'll find several the foomatic drivers. You can setup a printer name for each driver if you want ... one may be better for text, the other for colour pictures. For apps that won't let you choose which (like KDE apps it seems), select one as default. If you can't work things out, I think Till lurks in the general forum at linuxprinting.org. Good luck. Dave. On 21-Dec-2000 Praedor Tempus wrote: > It seems to be both a printer driver issue AND a printer hardware > problem. > I built and installed the gimp-print 4.1.0 package and changed my epson > driver to that. Now I can print from konqueror and kmail but, again, > the > printing is partial - even though I have a new black ink cartridge > installed, > I am getting ONLY color - no black (this may be the hardware - I am > going to > have to take the printer out of its hole and do a close exam to see if > the > print heads are clogged). As for the driver - even though I have 720dpi > selected as default, the printout is more like 180dpi or 360dpi. > > The gimp-print driver works MUCH better than the native driver. > > Another thing is that the default page size keeps going to A4. I keep > selecting letter and saving and hitting OK but it insists on defaulting > to > A4. I changed this as root to letter too - through kups and qtcups. > > How do I make letter the default that will stick? > > On Wednesday 20 December 2000 10:21 pm, you wrote: >> I'm running the stock konqueror from Enlightenment with the kde libs >> installed all from LM 7.2. (KDE 2.0.x). I have the same upgraded CUPS >> as >> you. I can print from Konqueror just fine. >> >> If you're printing off the page, your CUPS print driver is probably >> misconfigured. From the kups utility you can configure the driver with >> your paper size, resolution etc. You might also try a different driver >> ... >> each printer has several choices. Also Till setup a site with upgraded >> drivers which won't be part of Mandrake updates as they are not >> security >> related. I've had no need to try these however. >> >> Dave >> >> On 21-Dec-2000 Praedor Tempus wrote: >> > I am running KDE 2.1, cups-1.1.4-7.1mdk >> > >> > I thought I had cups setup properly for my Epson Color Stylus 600. >> > I can print out test pages and I just succeeded in printing (sort of) >> > an >> > email from kmail - sort of because it all ran off the top of the page > > > -- > Against stupidity, the gods themselves contend in vain > --- > praedor - 21-Dec-2000 13:39:54 -
RE: [expert] cups not working properly
I'm running the stock konqueror from Enlightenment with the kde libs installed all from LM 7.2. (KDE 2.0.x). I have the same upgraded CUPS as you. I can print from Konqueror just fine. If you're printing off the page, your CUPS print driver is probably misconfigured. From the kups utility you can configure the driver with your paper size, resolution etc. You might also try a different driver ... each printer has several choices. Also Till setup a site with upgraded drivers which won't be part of Mandrake updates as they are not security related. I've had no need to try these however. Dave On 21-Dec-2000 Praedor Tempus wrote: > I am running KDE 2.1, cups-1.1.4-7.1mdk > > I thought I had cups setup properly for my Epson Color Stylus 600. > I can print out test pages and I just succeeded in printing (sort of) an > email from kmail - sort of because it all ran off the top of the page > and > didn't print all the message text - all it produces is the KDE graphic > and > the From and To email addresses. Message body text isn't produced. > > In spite of this very limited "success" printing, I find it completely > impossible to print from konqueror. I tried repeatedly to printout a > webpage > I was reading but nothing ever comes of it. I checked the cups > error_log > file and it contains: > E [20/Dec/2000:21:20:20 -0700] PID 16396 crashed on signal 11! > E [20/Dec/2000:21:20:20 -0700] PID 16397 stopped with status 1! > > This is produced each time I try to print from konqueror. > So, cups isn't working properly. How does one fix this? Every time I > attempt to print from konqueror, konqueror freezes up for about 30 or 40 > seconds before cups gives up. > -- > Against stupidity, the gods themselves contend in vain > --- > praedor - 21-Dec-2000 00:21:03 -
[expert] CUPS + LM 7.2 install/config comments
Hi all (hope your there too, Till) After an LM 7.2 install/print struggle, I'd like to pass on a few printer related 'gotcha's' that I believe are problematic. I'll try and be as terse as I can (I have trouble being terse). Comments are most welcome. #1: /dev/lp0 vs /dev/lp1 When I upgraded from redhat 5.2 to 6.2 there was a big deal about how *now* the first parallel port is lp0, in lieu of lp1. I seem to remember the reason being kernel related, but it may have been my own redhat == linux subconscious. During my install of LM 7.2 I automatically chose lp0 as the parallel printer port and the install did *not* see the printer. Moreover, the CUPS documentation suggests that, in general, parallel printers are on lp1. Also, the "add printer" wizard from the kups tool still tells me I don't have a printer (it's on lp0 and working great now), and the wizard's parallel port list only shows "parallel port #1". I suspect both the LM 7.2 install routine and the kups wizard are broken in this respect, and if this is the case, then the many people fleeing from the redhat 7.0 mess to Mandrake might have some serious headaches. #2: CUPS http://localhost:631/adm Although this tool does see lp0, I suspect it will work if and only if you have a good /etc/cups/cupsd.conf file. At the moment, this http://localhost:631/adm does *not* work here, although I managed to get it to work very briefly during a blitz 'trial & error" session. I think it has to do with the "broadcasting" entries in the cupsd.conf file (see below). Nonetheless, this tool, which can configure your cupsd.conf file, will be useless unless you have a working cupsd.conf file to begin with. If I'm mistaken here, please let me know, because my default cupsd.conf file from the LM 7.0 install had all the important entries remmed out. #3: The /etc/cups/cupsd.conf file / I think this file could use a few more comments, or be fully docummented elsewhere. CUPS is new, and a 'man cupsd.conf' assumes a degree of knowledge (as it should) that was not needed with the admittedly 'brain-dead' LPD. After 4 years of battling with redhat, I like to think that I'm no longer a total Linux newbie. But, like many Linux users, I don't administer a multi-machine network, and the occasional comment relating to a single loopback system would be very helpfull. "Should broadcasting be on? Should I be broadcasting to 127.0.0.1? The comments certainly do not have to be comprehensive; a *single* sentence aimed at the home user who is not on a LAN, nor a LAN administrator, would be appropriate, I think, since a whole lot of mandrake/CUPS users are one home machine. Of course, that is what the wizards are for. But if the wizard is broken, or if it assumes underlying knowledge of the system, then that wizard is worse than useless. I think the kups "add printer" wizard is both broken and dangerous: you need to be lucky for it to work. Dave. (and aren't you glad I wasn't verbose?) Keep in touch with http://mandrakeforum.com: Subscribe the "[EMAIL PROTECTED]" mailing list.
RE: [expert] Replicated messages have multiple causes... PLEASE
So Pierre, why the hell are you posting twice? OOF. Bad joke, sorry! I'm sure everyone here greatly appreciates and *needs* your troubleshooting: we're suffering from some serious schizophrenia. A concensus might be that anyone posting from M$ Lookout! should be banned for life (at least). I was wondering what sort of delays people are getting in their posts. I'm in Montreal and, although I've only joined the list since the rather disastrous weather hit, my posts don't come back to me for 1 - 3 hours! I've become accustomed to 30 sec - 1 minute on other lists. Dave. On 19-Dec-2000 Pierre Fortin wrote: > I have analyzed just over 250 messages from Cooker and Expert lists. Of > those, > there were 20 duplicated and 3 triplicated messages. > > NOTE: If you are receiving this message from both the list and > DIRECTLY, you > should investigate your mailer, settings and/or habits for one or more > of causes >#2 and up since you were discovered in this sampling as being a possible > contributor to the problems. > >>From my short sampling, these are the causes of message replication I >>came up > with... > > CAUSE 1: "user 500"@yavin.mandrax.org: You should correct your parsing > to > distinguish between "To:", "Envelope-To:" or any other "*-To:" (anchor > regex to > line start, or stop processing headers) so as not to create more copies. > Also, > is there any chance your parsing is triggering on any other headers...? > Note: "*-To:" was added to the list at the last moment because as I > wrote this > message, a duplicate (see 2b) arrived containing "Reply-To:" as the only > discernable clue to the duplication. > > > CAUSE 2a: sender is using M$ Outlook Express configured to send an > "Envelope-To:" header. This is possibly one of the causes of the > duplicates; if > so, removing this header might help. Alternatively, lower-casing the > "T" in > this header might be another solution, if that's possible. The correct > solution > is most likely in Cause 1. Not every user of M$ Outlook Express sends > this > header, so it must be configurable. > > > CAUSE 2b: sender is using "Internet Mail Service (5.5.2650.21)" with a > "Reply-To:" header. See cause 2a for solution. > > > CAUSE 3a: sending to more than one addressee. > Examples: > To: [EMAIL PROTECTED], , [EMAIL PROTECTED] > To: [EMAIL PROTECTED], , [EMAIL PROTECTED] > > > CAUSE 3b: sending to both "To:" and "Cc:" > Example: > To: [EMAIL PROTECTED] > Cc: [EMAIL PROTECTED] > > CAUSE 3c: sending with BCC: which is not detectable from the messages > we get. > > > PLEASE, PLEASE, PRETTY PLEASE make the necessary corrections to > whatever > piece you control so that we can all maximize our enjoyment of these > lists. > > Hoping this makes a difference, > Pierre > > PS to yavin owner: any chance the above causes 2a-3c can also be > accounted for > in your "user 500" code...? > Keep in touch with http://mandrakeforum.com: Subscribe the "[EMAIL PROTECTED]" mailing list.
Re: [expert] CUPS exasperation
Yes, thanks go to John. I posted to linuxprinting.org and got a even better solution from Till. The WP driver for the Tektronix Phasor 140 is fully compatible with all the CUPS drivers. Apparently the postscript passthrough is not colour friendly. Refer to: http://www.mandrakeforum.com/article.php3?sid=20001129095531 Dave. On 19-Dec-2000 Kelley Terry wrote: > On Monday 18 December 2000 03:02 pm, you wrote: > >> > Hi, >> >> Although my setup is not exactly the same as yours, I had some similar >> problems, such as the cupsd stopping as soon as you try to print >> something. >> One thing I found is that in some cases, the program you are printing >> from >> may be set to provide output in the raw printer format, but cups >> expects >> the output in postscript and tries to generate printer code. The >> solution >> may be to set up Wordperfect to print "passthru postscript" and set the >> destination to be the printer you set up in cups. I just tried this on >> a >> remote Windows printer thru Samba and it works. > > > > Thank you! I have been trying to figure this one out myself and you > just > solved it for me! > Keep in touch with http://mandrakeforum.com: Subscribe the "[EMAIL PROTECTED]" mailing list.
[expert] CUPS exasperation update
In reply to myself, Been doing some debugging; The kupsdconf utility was not writing to /etc/cups/cupsd.conf. Almost everything was remmed out including ServerName etc. After some manual editing I have partial success. Problem: If I try to print from Wordperfect, it issues a command like: lpr -Pdeskjet520 /tmp/_pp_005957_8 (as listed from ps -aux) This kills the cups daemon instantly. The cups log ends with: cupsd[5643]: REQUEST localhost - - "POST / HTTP/1.1" 200 77 and a ps -aux shows the daemon as gone. No further errors are logged (and I'm in debug mode). What does work is printing from Nedit. It issues a: lpr -J "file.name" , although what shows up with ps -aux is: parallel:/dev/lp0 6 sundance file.menu 1 perl /usr/lib/cups/filter/cupsomatic 6 sundance file.menu 1 etc. So, does anybody have Wordperfect printing or any idea what might be causing it to kill cupsd? Thanks Dave. - 18-Dec-2000 16:37:21 - Keep in touch with http://mandrakeforum.com: Subscribe the "[EMAIL PROTECTED]" mailing list.
[expert] CUPS exasperation
Hi I really need to be able to print from Wordperfect 8/Linux real soon now or I'm sunk. (essay deadlines etc) With LM 7.2 Cups 1.1.4-7.1 local HP Deskjet520 on /dev/lp0 stand-alone dial-up system, with loopback working. I just can't get cups working properly. The kups program says can't connect to server. I've managed to configure my deskjet via the web config tool and the test page printed okay. I was also able to print from nedit. But the setup just won't stick. The daemon just dies after a very short time and my printer config is gone. Running chkconfing --list shows cups on at the appropriate run-levels. If I run "service cups start" or "/etc/rc.d/init.d/ cups start" I get an OK but if I immediately run lpinfo -v I get "Unable to connect to server: connection refused. All this is being done su'd to root with printer always on. Running kupsdconf as root I get: # QObject::connect: No such slot CupsdServerSecurityPage::cleanupEventFilter() QObject::connect: (sender name: 'unnamed') QObject::connect: (receiver name: 'unnamed') I just spent the whole day reading all the docs, and trying everything possible. I had updated cups using drakupdate and thought that may have been screwy so I downloaded the three rpms, uninstalled cups (rpm-e --nodeps), cleaned up the leftover directories, re-installed, and SNAFU. I'm now completely out of ideas and out of steam. Used redhat for 4 years and I'm starting to miss the old lpd. :-( Thanks in advance for any help. Dave. - 18-Dec-2000 04:12:39 - Keep in touch with http://mandrakeforum.com: Subscribe the "[EMAIL PROTECTED]" mailing list.