Re: wtmp locking problem (was: Re: SOLVED: Erk! Something is *really* wrong here!)
> Now this IS the answer. I just downgraded one of my systems to sysvinit > 2.69-1, rebooted, and the corruption seems to be gone. Few weeks ago, my corruption occured with sysvinit-2.69-1. I have been having this same package since December 8. For some unknown reason, few weeks ago my wtmp _stoped_ getting corrupted. And I have one dpkg request: An option that tells which packages were intalled since a certain date, i.g., % dpkg -older 2/28/97 . -- Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server.
Re: Security Packages
Last I looked, cops was not debianized. There is little reason for it since it need _lots_ (and simple) twinking to make it work right but with minimal results. It is not worth it. For filesystem security, it is worth to redirect your time to cfengine and tripwire (both are debian packages), to fixperms, and scanning regularly the log files. -- Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. On Tue, 18 Mar 1997, Ken Gaugler wrote: > Seems like there used to be a COPS package, or something similar, > that did a fairly automated security audit of one's Linux system. > > Can't seem to find or recollect what it was. Anyone know of a > program to do this? Is it a Debian package? > > Thanx! > > -- > Ken Gaugler N6OSK Santa Clara, California > email: [EMAIL PROTECTED] URL: http://www.wco.com/~keng > "The life of a Repo Man is always INTENSE..." > >
Re: lastlog is looking strange
A month ago wtmp was corrupted, today wtmp mysteriously is working fine. My experiences contradict those posted in this list: 1. I use ssh with no ill affects on wtmp. 2. mgetty was not installed on my system for over a year. While corruption lasted I was using getty. 3. Long uptimes does not seem to produce negative affects. My uptime today is 19 days and have no wtmp corruption. Since last month, I have probably installed less than 15 packages from bo. It looks like this problem is somehow corrected and will disappears as new packages get installed. Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. On Fri, 7 Mar 1997, Craig Sanders wrote: > > On Wed, 5 Mar 1997, Tim Sailer wrote: > > > In your email to me, Joey Hess, you wrote: > > > > > > Robert Stone: > > > > so am i the only one getting this? > > > > > > I'm getting it on two machines now. Both use ssh, if it's relevant. > > > The second is really bad: > > > > Bingo!! That seems to be the common factor. If you can reboot, do so. > > The problem seems to go away except for the accounts that are using > > ssh. My logins look like: > > > > tps tty2 *@ Sun Mar 2 23:56 - 02:26 (02:30) > > > > while others that have no ssh files look normal. > > I use ssh on all of my machines, and i use it regularly (instead of rlogin > and rcp and rsh). > > I don't get this wtmp corruption on most of these machines. I get a lot > of corruption on machines which have modems attached and a lot of modem > logins, a little corruption on machines with only a few modem logins, and > no corruption on machines with no modems. > > i initially suspected the new mgetty 1.1. however, i've just downgraded > to mgetty 1.0 and the corruption is still happening. So, the modem logins > may be completely irrelevant. > > Next on my suspect list is ssltelnet - i remember that weird things > happened in the utmp (but not wtmp) file just after i first installed > ssltelnet. i'm going to try reverting to normal telnet for a while and > see if that makes any difference. > > craig > >
Re: gzip repair
>> Martin> Not necessarily so. Perhaps "gunzip figures.tar.gz" >> Martin> followed by "tar -xvf figures.tar" works out all right. >> I have tried that, I get a "unexpected EOF error" when gunziping With very large files, 500+M, used to get bad gzips all the time. I never trusted it, and always had to check the new .gz with gzip -l . Do not know how gzip performs now. Since I back-up with dump(1). Try to use bzip ( not sure about the name, but it is in non-free). It might perform better than gzip. Or, try to buffer(1) the data before gzip sees them, maybe that will help. I don't think Debian has buffer(1), but you can get it from a sunsite mirror. It would be nice to know if Debian has a program that only buffers, anyone knows? Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. On Fri, 7 Mar 1997, root wrote: > > > > Maybe `gzip -dc figures.tar.gz > somename.tar` (and tar afterwards) will do > better? -- just a thought... > > Dimitri > >
Re: kerneld activity
On Wed, 5 Mar 1997, G. Kapetanios wrote: > I have posted this query previously but I got no replies. That is not quite correct. I did send you private email with suggestions to help you track the problem. That must have been within 24 hours since you first posted. If you do not know more about your problem it is because you did not try. Please refrain from suggesting that "you received no replies". In that message I suggested that you compile the kdstat program in /usr/src/modules-2.0.0/kerneld with "make kdstat". Then see what is happening by using the command kdstat debug, and kdstat nodebug to turn it off. Then I suggested YOU file a bug report using bug(1). It is not true you have been ignored. According to my logs, I responded within 24 hours: >>>>>>>>> From: Ioannis Tambouras <[EMAIL PROTECTED]> To: "G. Kapetanios" <[EMAIL PROTECTED]> Date: Thu, 27 Feb 1997 02:39:30 -0500 (EST) Subject: Re: kerneld activity <<<<<<<< >I am trying again because I think that this might be a bug. > Some days ago I noticed > after mounting an nfs system that the kerneld process would spawn other > kernelds all the time ( sometimes also sh processses ) and then kill them > only to do it again . I didn't pay much attention to it. This however > happens also when I mount ncp systems with the ncpfs program . Your original posting did not mention ncpfs. > When I > unmount the systems the activity stops. I have got most of my base files > from unstable but I only have Linux kernel 2.0.0 Could that be the cause > ?? > > I would appreciate any replies. Something else too. >... >... > George Kapetanios > Churchill College > Cambridge, CB3 0DS > U.K.E-Mail: [EMAIL PROTECTED] > Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server.
Re: Documentation - I see squares
The more(1) pager produces correct results when 8-bit characters are send to the screen. The less(1) pager fails to do so with its default configuration: I used "setenv LESSCHARSET latin1" to get around this problem. Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server.
make config, [486] or [586]
I have a AMD586-133 chip whose architecture more resembles an enchanced 486 cpu rather than a pentium. In terms of performence, is it better to compile with the [486] or the [586] parameter during make config? Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server.
Re: Documentation - I see squares
It might have to do with striping the 8th bit of a byte before it gets displayed on the terminal. For man, try the -7 option . That might not help you at all, but since the other responses from the list has not fix your problem yet, that is what I would investigate until a better suggestion arrives. Also, you may want to try different fonts, who knows. Or, try different options to /usr/bin/setterm, it might be. And if you find out, please, post the solution. Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. On Mon, 3 Mar 1997, John Foster wrote: > In a lot of man pages, and some of the documentation in /usr/doc there are > there little squares or cryptic <$%^> thingees. I guess that there's > something I've missed somewhere... > > What have/haven't I done? > > John. > > >
Debian Logo (joke)
Two foreigners, George and Nick, are visiting the white dominion of North Pole. This conversation takes place with one Eskimo in his igloo: George: Tell me, my good friend, are there white women in North Pole ? Eskimo: Of course there are. George: And how about black women, are there any? Eskimo: Very few women are black. But, why do you ask? George: How about mixed women, that appear in black AND white, are there any living here ? Eskimo: No strangers, there is no such thing. Then George turns and tells Nick: - And I told you yesterday, we were screwing with penguins... Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server.
Re: can't install TeX (fwd)
Take a look at the bug archieves for textbin at <http://www.debian.org/Bugs/db/>. One of your errors must be that texbin actually depends on mfbasfnt. Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. On Sun, 2 Mar 1997, Eugene H. Sevinian wrote: > Hi, > I will be thankfull for any advice to solve this problem. > > This is a part of the dselect's output. > I have found that 'libXt.so.6' is a link to 'libXt.so.6.0' in > /usr/X11R6/lib. > > Setting up texbin (3.1415-5) ... > kpathsea: Running MakeTeXTFM manfnt.tfm > Running MakeTeXPK manfnt.tfm > mf \mode:=nullmode; mag:=1; scrollmode; input manfnt \ mf: can't load library 'libXt.so.6' > Metafont failed for some reason on manfnt.tfm > kpathsea: Appending font creation commands to missfont.log. > dpkg: error processing texbin (--install): > subprocess post-installation script returned error exit status 1 > dpkg: dependency problems prevent configuration of latex: > latex depends on texbin (>= 3.1415-5); however: > Package texbin is not configured yet. > dpkg: error processing latex (--install): > dependency problems - leaving unconfigured > dpkg: dependency problems prevent configuration of ltxtool: > ltxtool depends on latex; however: > Package latex is not configured yet. > dpkg: error processing ltxtool (--install): > dependency problems - leaving unconfigured > Errors were encountered while processing: > texbin > latex > ltxtool > -- > > Regards, > Eugene Sevinian > > > Cosmic Ray Division > Yerevan Phisics Institute > Alikhanian's Brothers str.2 > 375036 Yerevan 36 > Armenia > > URL: http://www.yerphi.am/crd/prs/sevinian.html > Phone: 374-2-352041 (YerPhI), 374-2-344873 (aprt.) > Fax: 374-2-350030 > >
Re: shared library tutorial?
The elf how-to briefly mentions something about this, it says: "just compile all the object files with -fPIC, then link with a command like gcc -shared -Wl,-soname,libfoo.so.y -o libfoo.so.y.x *.o If that looks complex, you obviously haven't ever read up on the equivalent procedure for a.out shared libraries, which involves com- piling the library twice, reserving space for all the data you think that the library is likely to require in future, and registering that address space with a third party (it's described in a document over 20 pages long --- look at <ftp://tsx-11.mit.edu/pub/linux/packages/GCC/src/tools-2.17.tar.gz> for details)." I think that is what you want. Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. On Fri, 28 Feb 1997, Dale Martin wrote: > Hello, > > I have built a PCCTS source package - PCCTS is the "Purdue > Compiler-Construction Tool Set" - it produces LL(K) parsers. I'm > using it in a project which I will eventually Debianize. The PCCTS > package is close to ready to upload, except it has some libraries in > it, and I would like to compile them as shared libraries and don't > know how. (I'd also like to use shared libraries in my own project.) > > Can anyone point me to an online reference on how to compile and use > shared libraries? Note that I'm also interested in the portability of > the solution - my project also is working with Linux/Alpha, and > Solaris machines... > > Thanks for any info! > > Dale > >
Re: my zipper is stuck
To view the containts do: $ tar -zvtf guavac-0.2.5-linuxelf-bin.tar.gz To untar it do: $ tar -zxpvf guavac-0.2.5-linuxelf-bin.tar.gz Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. On Wed, 26 Feb 1997, Seth Reinosa wrote: > > how do I uncompress a progam that says > guavac-0.2.5-linuxelf-bin.tar.gz > > <[EMAIL PROTECTED]> > <http://home.eznet.net/~seth> > Thanx > and may God Bless you > Seth R > > > -- > TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to > [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] > > -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Free Publicity via the RC5 Challenge
> Also, with linuxnet at approx 70MKeys/sec and debian at about 10, we > don't really have any chance of overtaking them. We could have overtaken them, for two reasons: 1. MKeys/sec is a misleading number. I was running 30 process on one computer in order to upload them with one ppp connection. Because the processes are many, all of them have an ultra low keys/sec count. Most users have ppp, I know other people who do run 16 processes at night. Users like me cause a low Keys/sec count. It is the keySpaces/day that really counts. linuxnet had only a 4.5 higher keySpace/day count. 2. We just started 3 days ago, that's not enough time to estimate our strength. People are still looking for client, others have not read the debian-user for 3 days. We were not yet ready to race. Once the race got intresting, I would have been motivated enough to masquerade three other computers. Users like me, are very important. We may not control 40 computers, but for sure, we will be searching for keys every day. Lots of "would", and "would", yes I know the song. I my opinion, linuxnet was in big trouble. Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Package configuration philosophy
> Debian should provide a nicer default for the prompt. Many people > take this things into account when deciding which distribution they > like best. The flag of Texas should be a good prompt. DOS can do that, you know! Sorry, it's been a long day. This .sig is multi-threaded. == plato ~ fuser -km ~ioannis/rc5-client-linux-i586 /home/ioannis/rc5-client-linux-i586: 568c 568e 569c 569e 570c 570e 571c 571e 573c 573e 574c 574e 683c 683e 684c 684e 685c 685e 688c 688e 689c 689e 690c 690e 1107c 1107e 1108c 1108e c e 1112c 1112e 1113c 1113e 1114c 1114e 1494c 1494e 1834c Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. t -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: [OFFTOPIC] rc5-race is running!
After analyzing the data, I conclude that we theoretically have a chance to beat linuxnet. Our speed has droped 25% compared with yesterday. Linuxnet is running at four times our present speed. Today I received about six personals from debian users asking about the ftp site, or with other rc5-race questions. I think we do have a decent chance, once we get to position #2 (in about ten days) then we can take another look at our strength. Is the rc5-race package out yet? Four people have requested it since yesterday! It is already a popular item, and ideally (and if it is possible) it should come with a crontab job to upload the result and get a new keyspace. I tell you, we could *easily* compete with linuxnet in two weeks. Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. On Wed, 26 Feb 1997, Karl Ferguson wrote: > Hi. > > You may all be pleased to know the following stats at 12:25am +0800 WST > >1 [EMAIL PROTECTED] 112709 115.41 72.82 >2 #root 36550 163.60 16.66 >3 [EMAIL PROTECTED] 12229 95.72 9.53 >4 [EMAIL PROTECTED] 11350 133.86 6.32 >5 [EMAIL PROTECTED] 10394 78.17 9.91 > > With a 9.91 average, we're going to be able to get to number 3 easily - > however, I'm not sure if we'll be able to get any further! In any case, if > you would like to get started at cracking please do so. > > Regards > > > -- > ___ > >Karl Ferguson, >Tower Networking Pty Ltd [EMAIL PROTECTED] >t/a STAR Online Services [EMAIL PROTECTED] >Tel: +61-9-455-3446 Fax: +61-9-455-2776 http://www.star.net.au > ___ > > > -- > Please respect the confidentiality of material on the debian-private list. > TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to > [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] > > -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: [OFFTOPIC] rc5-race FAQ
1. DO I NEED PERMANENT CONNECTIVITY OR DIALD TO RUN THE CLIENT ? No. When the client starts, it establishes a tcp connection to zero.genx.net and gets a keyspace block, then it closes the connection. Once the key search is done, it will use a another tcp connection to return the result and get a new keyspace. There is no reason for internet connectivity during keyspace search. When the key search has finished and ppp is not on, the client will initially sleep for one minute and retry. After several times of failed attempts to reach the host zero, it will increase sleeping time to 10 minutes. 2. HOW MUCH BANDWIDTH ? --- It takes about 1k bytes or traffic per keyspace. For a P-133 that is about 1k every 30 minutes. 2. useful URLS More info about the contest is at http://zero.genx.net/ Email statistics are at http://zero.genx.net/bill/email.html RC5 clients are at ftp://portal.stwing.upenn.edu/pub/rc5 (US only) There are non-us sites with clients at (?? please let me know) 3. WHY NOT HAVE A DEBIAN PACKAGE ? Yes, we want to test one! 4. WHAT HAPPENS TO THE $10,000 PRIZE ? It is not clear who will declared the winner. Maybe it is top email address, or maybe the address that finds the key. Do not know. Debian has no official announcement on what will do with the prize. They should. Most likely, part or the whole prize will get donated somewhere. (where?) 4. WHAT ARE THE CLIENT COMMANDS ? -- To get a keyspace, use something command like this: % nohup rc5-client-linux-i586 -i [EMAIL PROTECTED] The -i option is needed to register the run to debian.org OR, to avoid any noticeable delays % nohup /usr/bin/nice -15 rc5-client-linux-i586 -i [EMAIL PROTECTED] % rc5-client- -m gives a rough idea how long is takes to complete one keyspace. For a P-133Mz it takes about 30 minutes, a 486-133 takes about 46 minutes. 5. ANY TIPS? -- On the first try, start with one keyspace and you will see the results in 30 minutes. The nice(1) is useful prevent noticeable delays. How many keyspaces you want depends on how soon you plan to reconnect to isp: with 16 keyspaces times 30 minutes per keyspace, you are done approximately 8 hours later. When you reconnect, all the 16 keyspaces will register at once. And if you forget to reconnect on time, the processes will sleep, and every minute will check for route connectivity. A crontab job comes handy. I found it it is inconvenient to manage more than 6 keyspaces at a time. If you want more, let 10 minutes elapse then get another 6 keyspaces. This makes it easier to figure out who is about to finish and who just started. And do not kill the wrong ones. Use renice(1) as root to undo what nice(1) does, if you have to. If you run high loads, say 20 or more keyspace searches, fetchmail will be unable to deliver to smtp. For sendmail, change the sendmail.cf to a higher value: # load average at which we refuse connections #O RefuseLA=12 Needs to be: O RefuseLA=30 Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: [OFFTOPIC] rc5-race is running!
On Tue, 25 Feb 1997, Hamish Moffatt wrote: > I'd consider running it here (since I have a permanently connected > machine with time on its hands), but how much network bandwidth does > it use -- it's at a premium here. Including inet headers, the upper limit is about 1k bytes of traffic per keyspace block. It all depends on how fast you finish the keyspace block, a pentium 133Mz pentium will traffic about 1k bytes every half an hour. Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
RC5: update
The score, as of Tue Feb 25 2:40 EST 1997, was: 1 [EMAIL PROTECTED] 100602 106.06 70.73 2 #root 35303 154.25 17.07 3 [EMAIL PROTECTED] 12063 86.37 10.41 4 [EMAIL PROTECTED] 10789 124.51 6.46 5 [EMAIL PROTECTED] 9964 124.50 5.97 6 [EMAIL PROTECTED] 8211 68.82 8.90 We are moving very fast to the top. Like super pacman, we are consuming 400 key blocks per hour, which is 8,000 key blocks per day. Two hours later debian will reach to top 5 and tommorrow to top 3. The stats of linuxnet and #root describe the contest: * linuxnet is untouchable and moves very fast. * #root runs at half our current speed. Their 35,000 blocks presently blocks our way, and will remain so for the next six days. Then we move to the number 2 position behind linuxnet. If things remain as they are, only linuxnet is a major player in this contest. Essentially, the same thing like three days ago before we got started. Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: [OFFTOPIC] RC5 Challenge
While spreading the news about the RC5 contest, I was asked the following question about the 56bit key and I wonder how to answer: What is the complication of this exhaustive search? I suppose it is exponential, but to what number? Thanks is advance Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: [OFFTOPIC] rc5-race is running!
Last night, I added one 133Mz to the effort. Here are few simple tips for those with dial-up isp accounts and with similar machine: (1) Connect to isp and get several Keyspaces by excecuting this several times: % nohup /usr/bin/nice -9 loop.sh rc5-client-linux-i486 -i [EMAIL PROTECTED] & (2) Disconnect from isp (3) When the all the Keyspaces are done, reconnect to isp and you will get more. There is plenty for everybody. :) The nice(1) is usefull prevent noticable delays. How many Keyspaces you want depends on how soon you plan to reconnect to isp: with 16 keyspaces times 30 minutes per Keyspace, you are done approximatly 8 hours later. When you reconnect, all the 16 keyspaces will register at once. And if you forget to reconnect on time, the processes will sleep, and every minute will check for route connectivity. A crontab job comes handy. On the first try, start with one keyspace and you will see the results in 30 minutes. I heard of a $10,000 award for the domain that comes firsts. That is why it is important to supply the "-i [EMAIL PROTECTED]" option. Thanks Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Sorry, again... for the RC5 stuff
I got mine from ftp://portal.stwing.upenn.edu/pub/rc5 If only I had masquerade for the other 4 computers, damn.. Now, only one computer eats on the tacos. Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: debian-newbie list
>> I'd like to see a standard for support questions that has people put >> keywords in the subject line. No need for such complications. Debian-user is fun, and I like the the subject lines the way they are. Sure, some answers are repeated, so what! That is part of the fun, the audience guesses at the questions once more and some day the ansers will settle, like jeopardy. Do not see anything wrong here, except (of course) for the deselect-ftp, the crond, the clear screen, the Netscape, and the syslogd questions. They are the ones that never end. No big deal... If the user wants to scan the archieves, he scans the archieves. He wants to post, he posts. He wants correct responses, he types $ man . He wants his answers posted, debian will post them. Who is willing to read long lists on the web? If I must browse through lists, I would rather read the long package manuals. We already have faqs, howtos, min-howtos, /usr/doc, nag, uag. You don't want to read anything, by all means, POST! ( in the-fastest-urls-are-my-D-and-ENTER-keys mode. ) Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: no NFS mounts from other hosts (!)
> You should uncomment the lines to start the nfsd and mountd on the other > hosts in the file /etc/init.d/netstd_nfs. Well, not if the only thing I want is to mount volumes. In that case you only need rpc support from the kernel. That is all. Thanks for all who replied. I will do some additional checks on the rpc calls with tcpdump to find why mountd failed to register with portmapper, then I will file a bug report. Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: clock
If you set the system clock (the real clock) to GMT, then MS Windows will also think that your are in England. (Maybe in Windows95 has a a way around it, don't know). To do that, (most likely you do not want it!) supply the "-u" to the timezone script. It should work, but I never tried it. Alternativelly, set the kernel clock manually with the date command, then " clock -w" to write it permanetly on the system (hardware) clock. You see, there are 2 different clocks: 1. is what the kernal thinks and manipulated with date(1). 2. the real clock on the motherboard, manipulated with clock(1). Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: clock
> > I assume that you don't want the "-u" flag. > > So, how can one figure out (from DOS or the BIOS) if your system clock > is set to GMT or not? Other than trial and error... > > Rik. Your .be domain tell me that you are +1 hours ahead from Greenwich (sp?), England. Check the current time that dos tells you, and if it agrees with your local time then your system clock is not set to GMT. You can also check from linux at what time your system clock is set with % clock -r . If this output matches that of % date -u , then your system clock is set to GMT In other words, if c:> time , and % date show you the right time and your are happy with that, then you are all set. Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: no NFS mounts from other hosts (!)
[Jerry, allow me to send a copy to debian-user.] > mount -t nfs -o mountport=824 server:/world /world > > where 824 is the tcp proto port for mountd on the server. Now, it works - THANKS! The problem apparenlty is with the portmapper in netbase_2.06-1 in bo (unstable). HostA (was called "server" in my previous posting) uses both versions of mountd, ver. 1 and mountd ver. 2, and has no problems mounting from itself or from host B. Host B , the "client", uses mountd ver 1, and only mounts from itself. To mount from HostA, as you advised me, it needs the "-o mountport" option. Host uses netbase_2.06-1 (stable), whose corresponding netstd package provides only one version of mountd. I will wait 12 hours for any comments, then I will report a bug on the portmapper of the unstable netbase package. Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- This message was delayed because the list mail delivery agent was down.
no NFS mounts from other hosts (!)
NFS mounts for other hosts fail. On the server everything is in place: portmap, nfs, mountd, are running, and I can mount nfs partitions from the server to itself. But, from another host (who has nfs module loaded) I get the classic: "mount clntudp_create: RPC: Program not registered" What else should I check? On the server: % cat /etc/export /world (ro,root_squash) % rpcinfo program vers proto port 102 tcp111 portmapper 102 udp111 portmapper 132 udp 2049 nfs 132 tcp 2049 nfs 151 udp821 mountd 152 udp821 mountd 151 tcp824 mountd 152 tcp824 mountd % ps -xa (partial listing) 117 ? SW 0:00 nfsiod 118 ? SW 0:00 nfsiod 119 ? SW 0:00 nfsiod 120 ? SW 0:00 nfsiod 1066 ? S 0:00 /usr/sbin/rpc.nfsd 1068 ? S 0:00 /usr/sbin/rpc.mountd Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Debian review in Linux Journal
On Tue, 21 Jan 1997, Eloy A. Paris wrote: > I'd like to know if someone can give me a link to a place where I can read > the Debian review made by the Linux Journal staff in their November 1996 > issue. Do not think that LJ has all their articles on-line. They have some, but not the Debian review article. There is also a distribution comparison, dated Jan 15 1997, but it contains ancient information, http://www.ssc.com/lj/distable.html. Ioannis Tambouras [EMAIL PROTECTED] PGP 768/429EE365, West Palm Beach, Florida -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: How come no Aladdin ghostscript?
The gs-aladdin_4.03-6.deb package (or a newer version) is in non-free. Ioannis Tambouras [EMAIL PROTECTED] PGP 768/429EE365, West Palm Beach, Florida On Mon, 20 Jan 1997, Robert Nicholson wrote: > Isn't it regarded that the Aladdin ghostscript offers better > fonts/features? > > > -- > TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to > [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] > > > -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: bash question ...
> bash$ xplaycd & ; xmixer & > bash: syntax error near unexpected token `;' Indeed, sh(1) will breaks in svr4, but not in bsdi. Did not spend the time to track down the verisions thought. In debian, bash(1) version 1.14.7-2, this problem is easily solved, do: $ xplaycd & xmixer & Do not know about trap differences. Ioannis Tambouras [EMAIL PROTECTED] PGP 768/429EE365, West Palm Beach, Florida On Sat, 18 Jan 1997, Richard G. Roberto wrote: > Are there any bash gurus out there? I have a couple of Q's. > > I have some scripts that run fine under bourne-shell on solaris, > but break under bash on debian. The signal trapping doesn't seem > to work. Nor does executing multiple commands on a single line > separated by ";" characters. Even on the command line: > > bash$ xplaycd & ; xmixer & > bash: syntax error near unexpected token `;' > > Any help would be appreciated. > > Thanks > > Richard G. Roberto > [EMAIL PROTECTED] > 011-81-3-3437-7967 - Tokyo, Japan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Modprobe. Two of a kind ?
Is it possible to load two ne.o modules with different io= options ? This computer is multihomed to two ethernets. I like to be able to configure the interfaces like this: eth0 with ne.o io=0x240 , interupt 15 and eth1 with ne.o io=0x280 , interupt 12 Already know how to do it with lilo when the code is an intergral part of /vmlinuz, but does not work when everyting is modularized. That is not covered in Becker's mini-HOWTO. I played with the modprobe(8) options and aliases, but nothing looks promissing. Ioannis Tambouras [EMAIL PROTECTED] PGP 768/429EE365, West Palm Beach, Florida -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Any PPP-connection keeping utility ?
On Mon, 6 Jan 1997, Lawrence Chim wrote: > > I put this in my /etc/ppp/ip-up script: > > > > ping -i 120 $5 > /dev/null 2>&1 & >^ >| >what is it? not found in man page > The $5 represents the ip address on the isp side. >From the /etc/ppp/ip-up comments: #Arg Name Example #$1 Interface name ppp0 #$2 The ttyttyS1 #$3 The link speed 38400 #$4 Local IP number12.34.56.78 # $5 Peer IP number12.34.56.99 Ioannis Tambouras [EMAIL PROTECTED] PGP 768/429EE365, West Palm Beach, Florida -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: ping reply without OS
Having read that the interface card is capable to reply to icmp echo requests, my next question should be this: Why does does the card assume that we are using the internet protocols? (On a side note, this does no good to those who ping the machine and conclude that everything is well with the ip layer of the halted machine, when in fact it should be dead.) My real complain is when I read "System halted", my assumption has always been that the cpu has executed the x86 HALT instruction . Why not, it consumes a lot less enery in this state. Ioannis Tambouras [EMAIL PROTECTED] PGP 768/429EE365, West Palm Beach, Florida -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Unidentified subject!
GOOD YEAR everyone, After host plato is halted, I ping it from another host on the same ethernet. Plato responds with echo replies! Both hosts are on debian 1.2 . Maybe something is wrong with shutdown, but except for ethernet collisions I cannot think of any other consequences. Ioannis Tambouras [EMAIL PROTECTED] PGP 768/429EE365, West Palm Beach, Florida -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: icmpinfo (old thread)
I am responding to a request which was posted about a month ago... I can read the first 3 lines of each output, they represent two ip headers, (icmp errors reflect the ip header which caused the error back, as part of their reply). I do not know what the bytes after the first 3 lines represent, theoretically, these bytes are supposed to be the begining of the original data send with the original ip datagram (represend original udp headers) which caused the icmp error. That's theoretically, but I am only sure about the first three line. The first icmp port unreachable error are in responce to a udp(0x11) datagram send by localhost to localhost (7F00 0001 7F00 0001 ). The rest of the information shows, among others: id of undemultipled ip was 06FD id of ip of this icmp was 06FC (again, they were generated from the SAME host) The rest of the icmp port unrechable requests also are from localhost and for the same reason. > [EMAIL PROTECTED] Tue Dec 17 20:06:09 1996: > > I'm in need of a TCP/IP expert here to tell me if someone is > trying to spoof/ping flood me... I know someone has tried to 'big ping' me > several times due to the "couldn't get a free page" message on my console. > I've been running icmpinfo -vvv > /tmp/icmplog, and I'm getting alot of > ICMP_Dest_Unreachable messages. Is this normal? They're comming mostly > from localhost but also from other sites. Could someone please advise me > on what to do, or where to get some more info on how to find out where > these are comming from? Here are several of the 'pings' I've gotten. > > Nov 25 18:31:42 ICMP_Dest_Unreachable[Port] < 127.0.0.1 [localhost] > > 127.0.0.1 > [localhost] sp=25861 dp=53 seq=0x0033adea sz=79(+20) > : 4506 0063 06FD 4001 7595 7F00 0001[EMAIL PROTECTED] > 0010 : 7F00 0001 0303 FB43 4500 0047...CE..G > 0020 : 06FC 4011 75A8 7F00 0001 7F00 0001[EMAIL PROTECTED] > 0030 : 0565 0035 0033 ADEA 001A 0100 0001 .e.5.3.. > 0040 : 0136 0236 3103 3130 3203 32 .6.61.102.2 > > Nov 25 18:18:43 ICMP_Dest_Unreachable[Port] < 127.0.0.1 [localhost] > > 127.0.0.1 > [localhost] sp=17669 dp=53 seq=0x00360a1f sz=82(+20) > : 4506 0066 069F 4001 75F0 7F00 0001[EMAIL PROTECTED] > 0010 : 7F00 0001 0303 FB46 4500 004A...FE..J > 0020 : 069E 4011 7603 7F00 0001 7F00 0001[EMAIL PROTECTED] > 0030 : 0545 0035 0036 0A1F 000A 0100 0001 .E.5.6.. > 0040 : 0332 3332 0331 3432 0331 3931.232.142.191 > 0050 : 0332 .2 > > Nov 25 18:38:28 ICMP_Dest_Unreachable[Port] < 127.0.0.1 [localhost] > > 127.0.0.1 > [localhost] sp=33285 dp=53 seq=0x0034380d sz=80(+20) > > There was no data in the last entry to the file. The data of the > ping almost always seems to have an IP address in it. What can I do, or > am I being paranoid? > > TIA, > mike... > Ioannis Tambouras [EMAIL PROTECTED] PGP 512/D042DD45, West Palm Beach, Florida -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: moving directories to new partition
>I am in need of some advice about how to go about moving some of the >directories off of my main partition and onto one I just made. (1) from Tips-HOWTO: Quick way to move an entire tree of files from one disk to another (cd /source/directory && tar cf - . ) | (cd /dest/directory && tar xvfp -) (2) I saw a program in sunsite called "reflect". I think it is in /pub/Linux/utils/files Ioannis Tambouras [EMAIL PROTECTED] PGP 512/D042DD45, West Palm Beach, Florida -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Why does "stable" point to "rex"?
The symbolic links tell what is stable or unstable, not the dir names. When debian_1.2 was released few days ago, it was time to change the links to point at the right direction. So, the faq is now incorrect. Ioannis Tambouras [EMAIL PROTECTED] PGP 512/D042DD45, West Palm Beach, Florida -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
package creation, basics
After reading the debian-faq, I am having trouble creating a debian package for local use. One day it will reach debian.org too, I hope! % ls control.tar.gz data.tar.gz debian-binary control/ data/ % /usr/bin/ar -rc ascii-1.2_1.deb control.tar.gz data.tar.gz debian-binary This ascii-1.2_1.deb (my unofficial debian package) is not recognized by dpkg as a valid .deb package. What should I have done instead? Ioannis Tambouras [EMAIL PROTECTED] PGP 512/D042DD45, West Palm Beach, Florida -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: color-ls??
> I can't seem to find the color-ls package. Has it been replaced by > something else? No more color-ls. It is now part of fileutils, a base package, so it is already installed. Add few aliases to /etc/csh.login for the miracle to happen (for more info, see /usr/doc/fileutils/color-ls.gz): eval `dircolors` alias ls 'ls --color=auto '; alias ll 'ls -l'; alias dir 'ls --color=auto --format=vertical'; alias v'ls --color=auto --format=long'; alias ols '/bin/ls'; Ioannis Tambouras [EMAIL PROTECTED] PGP 512/D042DD45, West Palm Beach, Florida -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
ncftp. Current boot disks
I need clarification on two issues: * The symbolic link ftp://ftp.debian.org/debian/rex/disks-i386/current is pointing at 1996-11-28/, instead of 1996-12-7/. I think that is incorrect. * I need to file a bug report for ncftp-2.4.2: the get -R command is not excecuted, I only get the prompt for the next command. Few days ago I was looking at a debian bug list. Now that I need to check if this bug is reported, I cannot find it. The debian-faq.txt directed me to a debian-bugs/ archieve, but there was nothing there! Where can I browse for previous bug reports? Ioannis Tambouras [EMAIL PROTECTED] PGP 512/D042DD45, West Palm Beach, Florida -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: internet-time
% /usr/sbin/netdate clock.llnl.gov && /sbin/clock -w clock.llnl.gov was (and probably still is) a NTP stratum 1 server. I prefer it not for its accuracy, but because this host is always up and running. Good for scripts. This is the only server that I know. If someone knows where to get the list of stratum servers, please post! Ioannis Tambouras [EMAIL PROTECTED] PGP 512/D042DD45, West Palm Beach, Florida -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Duplicating Debian Installations?
$ umask 000 # not sure if it is nessesary $ (cd /; tar -clpf - .) | ( cd / ; tar -xpkf - ) $ umask 022 This is from the Tips-HOWTO, there is one other method in that doc but never used it. With the method shown above you might get some complains about not being able to copy some /dev devices (or was it just some named pipes). I always ignore those errors, and everything works just fine. Of course... Once you transfer, remember about: rdev, /etc/fstab, etc.. Ioannis Tambouras [EMAIL PROTECTED] PGP 512/D042DD45, West Palm Beach, Florida On Mon, 19 Aug 1996, Tres Hofmeister wrote: > > Can someone point me towards info. on how one can duplicate the > packages installed on one Debian system on another? Going through > dselect by hand seems just a bit too tedious for multiple > installations... Thanks.
Re: lost lib
On Tue, 6 Aug 1996, Robert Van Horn wrote: telos 16 % zgrep -i librl.so /data/deb/Contents.gz lib/librl.so.2.0.3.dist aout-librl It is in: filename: Debian-1.1/binary-i386/devel/aout-librl-2.0.3-4.deb msdos-filename: Debian-1.1/msdos-i386/devel/aoutlbrl.deb size: 112098 md5sum: 46972d5000aa7785bad36eed1d4505e7 > When I try to use ftp I get a message "ftp: can't find library 'librl.so.2'" > I also am not able to find this library any of the places I looked. > > Thanks in advance for any help finding this library. > > Bob > > > > Ioannis Tambouras [EMAIL PROTECTED] PGP 512/D042DD45, West Palm Beach, Florida
dosemu, libX11.so.6 problem
Trying dosemu-0.60.3-1.deb to run on linux 2.0.10 Compilation on .tgz sources also error and do not want to compile with X support, have not yet tried without it. Will appresiate any help. After dpkg -i the binary package: problem === $ dpkg -i x11/xlib-3.1.2-7.deb ... $ cp /usr/X11R6/lib/libX11.so.6.0 /usr/lib/libX11.so.6 $ dos -A dos: can't load library '/usr/lib/libX11.so.6' Exec format error dos: can't find library 'libX11.so.6' ======== Ioannis Tambouras [EMAIL PROTECTED] PGP 512/D042DD45, West Palm Beach, Florida
