A Question about the Application Called mail or mailx
Until I get out-bound messages going through nmh properly, I have found a possible stop-gap measure to use. The old mail application or mailx if one has heirloom-mail does work but I have a question about piping a message to it. It looks from documentation that mail can read headers such as: From: martin McCormick mar...@myhost.net To: mar...@testhost.com Subject: Will this ever work? that are embedded in a message file if one sends a command something like mail -t [filename] but what actually happens when I try this is an error stating a complaint that there are no recipients on the command line. That's right. I put them in the file so I shouldn't have to put them in the command line. Should this work and produce a sent message? Thank you. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150722174027.c5ed822...@server1.shellworld.net
Re: A Question about the Application Called mail or mailx
Dan Ritter writes: mail -t needs to be followed by an address, not a message body. Makes sense. Thanks. If you want to send a full message which has all needed headers, trust the sendmail command which is shipped by anything which can supply the MTA package role. exim, sendmail, postfix, qmail, masqmail... they all give you a sendmail command which will do what you want. In this case, it is msmtp which I call as sendmail and it worked like a charm. Many thanks for the help. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150723013004.cacff22...@server1.shellworld.net
Re: Fetchmail may almost be working in pop3 Big Progress
To all who have helped me so far, a huge thank you! I soon realized that the subject line of this message is incorrect since pop3 covers only the delivery task and I got that working a couple of weeks or so ago. The indescribably joyful experience of being able to successfully authenticate to an out-going smtp smarthost that my ISP provides, however, has more than made up for the false sense of mission accomplished I felt when the first pop3 message popped up. I did take the output of an unsuccessful authentication attempt and ran the base64 string it returned as the password through base64 in decode mode and, voila, there was my user ID and password in nice clear text as it should be and there's where I found the last unexplainable reason for failure. In .msmtprc, there is a line that looks like: password 'BiG-SeCreT' I wanted to be really concise so I enclosed the secret in single quotes just as you see it here. I expected msmtp to strip those out and that would insure that what was there was always taken literally. What's there is taken literally, all right including the single quotes. I took them out and I can now see that authentication is successful and a couple of test messages went out and have arrived at a remote host in 1 piece so I technically can send mail but I still need to tie msmtp in to nmh which some of you may be familiar with. That is a different topic for a different mailing list. Again, thanks to all for your patience. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150721163034.ef1c822...@server1.shellworld.net
Re: msmtp Questions
Bob Bernstein writes: At some point in this process try using your POP credentials (username martin, password martin's pop password) when trying to send. Y'know, when suddenlink told you that 'martin' had been already taken as a username for smtp, of course it had, by YOU, for your POP account. Just a thought. I'm just sayin'. Good idea, but I can answer the easiest question first. The name that originally setup this account is based on my wife's user ID. I first tried mar...@suddenlink.net and the system let me type it in then ate it a few seconds later and popped up a message saying that this user ID is unavailable. It did that same thing to attempts to register a couple of other UID's before liking martin.m. Those credentials work without fail when retrieving messages via fetchmail and one must use the same credentials with the smtp host as one does to retrieve mail. Just the same, you caught me on one thing. the full user name is the full email address and I was just using martin.m. Fixing it changed nothing. The following is a partial quote from Suddenlink's instructions to pop and smtp users. This is the part that applies to what doesn't yet work: SMTP (outbound) Outgoing Mail Server: smtp.suddenlink.net Outgoing mail port: 25 or 587 Outgoing mail port (SSL): 465 Note:Username: your full email address, including the @domain.net at the end (i.e. @suddenlink.net) Password: The password set to your particular email address Outgoing server verification must be enabled and on. If your email client has the option or checkbox that says My outgoing server requires authentication this will need to be checked along with use same settings as my incoming mail server. If the email client asks for Clear Text Authentication, check yes. It looks like you log in in the clear and then encryption is used to send the mail. If you start using port 465 or anything but 25 or 587, it just hangs until the next big bang or I get bored waiting. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150718111003.d938922...@server1.shellworld.net
Re: msmtp Questions
Curt writes: You don't seem to be following the instructions here: After a good night's sleep, I notice that too. I fixed it and now there are only moving parts, one of which is broken instead of 1.:-) Unfortunately, authentication is loaded with these series-connected lamps, any one of which makes the whole string dark. Thank you for noticing, however. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150718112334.ddbd022...@server1.shellworld.net
Re: msmtp Questions
David Wright writes: Supported authentication methods: PLAIN LOGIN I see no encryption here. I think this is why it is telling you that it cannot use a secure authentication method. I wrote You might want to check out port 587 but I think you'll be disappointed just this morning. As I wrote there, you should try port 465. I'd like to tell you it is working but you are right about the port. I tried your suggestion but turned off starttls this time as starttls turned on causes a short hang followed by a message stating that starttls won't work. I freely admit I do not understand the fine details of TLS other than it is a trust system based on certificates and you get end-to-end encryption of the session. Here's how this session went. I assume the text came encrypted with TLS. The lines now have a carriage return after them. Since I piped all output to a file, the standard error came in first. msmtp: authentication failed (method PLAIN) msmtp: server message: 535 Authentication failed msmtp: could not send mail (account default from /home/martin/.msmtprc) loaded system configuration file /etc/msmtprc loaded user configuration file /home/martin/.msmtprc using account default from /home/martin/.msmtprc host = smtp.suddenlink.net port = 465 timeout = off protocol = smtp domain= localhost auth = choose user = marti...@suddenlink.net password = * passwordeval = eval ntlmdomain= (not set) tls = on tls_starttls = off tls_trust_file= /etc/ssl/certs/ca-certificates.crt tls_crl_file = (not set) tls_fingerprint = (not set) tls_key_file = (not set) tls_cert_file = (not set) tls_certcheck = on tls_force_sslv3 = off tls_min_dh_prime_bits = (not set) tls_priorities= (not set) auto_from = off maildomain= (not set) from = marti...@suddenlink.net dsn_notify= (not set) dsn_return= (not set) keepbcc = off logfile = /home/martin/.msmtp.log syslog= LOG_MAIL aliases = (not set) reading recipients from the command line TLS certificate information: Owner: Common Name: *.suddenlink.net Organization: Suddenlink Communications snip From the server: -- 220 txofep01.suddenlink.net ESMTP server (InterMail vM.8.04.03.20 201-2389-100-164-20150330) ready Sat, 18 Jul 2015 06:45:58 -0500 -- EHLO localhost -- 250-txofep01.suddenlink.net -- 250-HELP -- 250-XREMOTEQUEUE -- 250-ETRN -- 250-AUTH=LOGIN PLAIN -- 250-AUTH LOGIN PLAIN -- 250-PIPELINING -- 250-DSN -- 250-8BITMIME -- 250 SIZE 52428800 The password -- AUTH PLAIN (encoded to end of line) -- 535 Authentication failed I see what msmtp means about the password. I bet the encoded string is the base64 password you mentioned. Anyway, thanks! I think we are inching (more like millimetering) forward. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150718121356.e78ea22...@server1.shellworld.net
Re: Fetchmail may almost be working in pop3.
Paul E Condon writes: I use msmtp, not exim, even though exim comes already installed by Debian. Msmtp has its own tiny config file which can be located at ~/.msmtprc You can put there whatever you need to satisfy you ISP and have no fear of exim mucking about with it. Of course, don't remove exim once you see msmtp working. That would break you Debian installation. Msmtp is a package in the main branch of all Debian repositories. Thank you as I believe this will do the job much more intuitively than exim4. Should I reset exim4 back to local delivery since it does still need to be able to handle squawks and status messages for cron jobs and other self-generated shreaks and howels (bells and whistles gone wrong). I never had a .mailrc file on this system before but it is going to have to know that sendmail is now msmtp. I looked at sendmail's link after installing msmtp and it still points to exim4 so nothing got changed system-wide. msmtp --serverinfo now gives me a banner page from smtp.suddenlink.net listing capabilities. Interestingly, they list SSL ports in their documentation but the server says it can't do starttls. The banner page came in on port 25. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150717144032.98cce22...@server1.shellworld.net
Re: IP address
Miles Fidelman writes: ifconfig -a is always a good one Yes but depending on how your path is set it may not simply work. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150717191057.cb60322...@server1.shellworld.net
msmtp Questions
The fun never ends. I installed msmtp and as near as I can tell it works as advertised. My SMTP smarthost at Suddenlink.net presents the following banner which nicely explains what one needs to do to get real work done. I've had a little trouble getting msmtp to fit what is required. The documentation provided by Suddenlink says that smtp is provided on ports 25 and 587 and both display the following banner: SMTP server at smtp.suddenlink.net (smtp.suddenlink.net [208.180.40.68]), port 587: txasav-vm06.suddenlink.net ESMTP server (InterMail vM.8.04.03.20 201-2389-100-164-20150330) ready Fri, 17 Jul 2015 19:59:42 -0500 Capabilities: SIZE 52428800: Maximum message size is 52428800 bytes = 50.00 MiB PIPELINING: Support for command grouping for faster transmission ETRN: Support for RMQS (Remote Message Queue Starting) DSN: Support for Delivery Status Notifications AUTH: Supported authentication methods: PLAIN LOGIN Here's the .msmtprc file with password redacted. # Set default values for all following accounts. defaults tls off tls_trust_file /etc/ssl/certs/ca-certificates.crt logfile ~/.msmtp.log # Set a default account account default auth on from marti...@suddenlink.net protocol smtp port 587 # The SMTP smarthost. host smtp.suddenlink.net user martin.m password 'IfItoldu,Idhave2' passwordeval eval # Construct envelope-from addresses of the form user@oursite.example. auto_from off # Don't use starttls. tls off tls_starttls off #tls_trust_file /etc/ssl/certs/ca-certificates.crt # Syslog logging with facility LOG_MAIL instead of the default LOG_USER. syslog LOG_MAIL When I run a test by piping a file in to msmtp, I get loaded system configuration file /etc/msmtprc loaded user configuration file /home/martin/.msmtprc using account default from /home/martin/.msmtprc host = smtp.suddenlink.net port = 587 timeout = off protocol = smtp domain= localhost auth = choose user = martin.m password = * passwordeval = eval ntlmdomain= (not set) tls = off tls_starttls = off tls_trust_file= /etc/ssl/certs/ca-certificates.crt tls_crl_file = (not set) tls_fingerprint = (not set) tls_key_file = (not set) tls_cert_file = (not set) tls_certcheck = on tls_force_sslv3 = off tls_min_dh_prime_bits = (not set) tls_priorities= (not set) auto_from = off maildomain= (not set) from = marti...@suddenlink.net dsn_notify= (not set) dsn_return= (not set) keepbcc = off logfile = /home/martin/.msmtp.log syslog= LOG_MAIL aliases = (not set) reading recipients from the command line -- 220 txasav-vm07.suddenlink.net ESMTP server (InterMail vM.8.04.03.20 201-2389-100-164-20150330) ready Fri, 17 Jul 2015 22:01:57 -0500 -- EHLO localhost -- 250-txasav-vm07.suddenlink.net -- 250-HELP -- 250-XREMOTEQUEUE -- 250-ETRN -- 250-AUTH=LOGIN PLAIN -- 250-AUTH LOGIN PLAIN -- 250-PIPELINING -- 250-DSN -- 250-8BITMIME -- 250 SIZE 52428800 msmtp: cannot use a secure authentication method msmtp: could not send mail (account default from /home/martin/.msmtprc) Believe it or not, this is the closest I have gotten to making anything work. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150718031040.4c86622...@server1.shellworld.net
Re: wheezy to squeeze
Steve McIntyre writes: mar...@server1.shellworld.net 622 upgraded, 0 newly installed, 0 to remove and 551 not upgraded. Need to get 222 MB of archives. After this operation, 48.4 MB of additional disk space will be used. Do you want to continue [Y/n]? Needless to say, I typed n and there's where things stand now. The number of held-back entries is about the size of the whole distribution so something is seriously wrong. What command are you using to upgrade? This looks like you're using apt-get upgrade when you need apt-get dist-upgrade ... After upgrading to wheezy, I ran apt-get update one last time followed by apt-get upgrade and then apt-get dist-upgrade. To complicate things, I discovered my sources.list file was missing the line deb http://ftp.us.debian.org/debian/ wheezy contrib non-free I added it and it did pickup a few more packages. I then did upgrade with no problem and then dist-upgrade to get the wheezy versions of anything found. It again worked without a complaint. It is after I replace all occurrances of wheezy with jessie and do the update, upgrade and dist-upgrade commands that it shows all the held-back packages. I do have the original squeeze system on a drive and my upgrade was started by copying the entire squeeze drive to this new drive. If worse comes to worse, I could do the whole thing over now that I have a sources.list file or better yet, I need to find a sources.list file that is sure to have everything it is supposed to have so this doesn't happen again. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150716175132.995ca22...@server1.shellworld.net
Re: the State of Linux Audio
Joel Roth writes: Hi Martin, Pulse audio requires D-Bus, and D-Bus is the underlying RPC mechanism of a large and controversial software stack developed to support desktop applications. Thank you for this good and quick explanation. Apparently pulseaudio is unable to get D-Bus services, due to a dependency of the latter on X. So you need either to satisfy/finesse this dependency of D-Bus, or disable/remove pulseaudio. I've read but not tested apulse, a library that purports to presents a pulse audio API to applications such as skype that require them, relaying the audio to ALSA. It looks like pulseaudio has been the ghost in the basement on my system for about ten years. For now, I took it off completely and killed the process. Before the CS4236 went away, there was always a sound device for Card 0 and, if I had a second sound card, there was another sound device for the second card. It was typical to see /dev/dsp linked to Card 0, an actual /dev/dsp0 device and /dev/dsp1 for Card 1 and so on. Just for fun, I think I stuck in a USB sound card in addition to cards 0 and 1 and predictably got /dev/dsp2 plus a lot of strange audio that sounded like a bad tape transport due to all the sound cards trying to write to their little segments of memory at the same time on a 600-MHZ Pentium. Back to the present, I still had PA (pulseaudio) running, no official Card 0 and a USB-based Soundblaster Digi acting as Card 1. At least that's what aplay -l said. I could run mplayer and the playback was excellent but amixer for Card 0 only showed one control for left and right front volume. After I killed PA, mplayer said it couldn't find any cards as there was no /dev/dsp any more. I did however find /dev/dsp1 for the SBdigi so I manually forced a link with ln -s to link /dev/dsp to /dev/dsp1. Presto! mplayer was happy again and played music and other audio files but the story isn't quite over yet. I'm thinking that pulseaudio does some signal processing, also. Some of the sound files I listen to are 8-bit PCM voice recordings made at 8000 samples per second. They're just fine for recording two-way radio chatter. Mr. Nyquist is happy because 3 to 4 KHZ is the highest frequency you will usually hear over such communication so it doesn't sound much different than it did when first heard over the air. Before I killed PA, the audio of those raw PCM recordings sounded fairly normal. After I killed PA, you can now still hear the audio but you can also hear the 8-KHZ sampling which sounds like a cheap toy as most of those don't bother with a low-pass filter but let your ears and brain do that. It is possible that pulseaudio is using DSP techniques to shape the wave forms properly and then is up-converting the samples that the sound card sees. You can't add any fidelity that is not already there, but this would act as a very good low-pass filter. I also got in to /etc/modprobe.d and added a line to alsa-base.conf to make the SBDigi be card 0 until I can resurrect the CS4236 and this seems to have made everything work automatically again. amixer now reports a full-featured sound card with all the controls one needs to do good playback and normal recording. The playback is actually better than the CS4236 was so now we have some progress. After things settle down, I may put PA back just for the signal processing but for now, it's best to keep things as simple as possible. Again, thanks for the explanation of some of pulseaudio's purpose in life. It is presently used in those screen reader modules that allow the kernel to generate synthesized speech so it isn't all bad but it sure helps to know what it does. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150716111542.77fb622...@server1.shellworld.net
/dev/dsp Obsolete or Not?
What replaces the standard sound device? I have written some experimental programs that play and record sound using /dev/dsp and they work. Obviously, there is a lot of bad design in the world that works and I hear the discussion that says that /dev/dsp is out-dated so what is considered the best way to send and receive audio in a program? If there is no /dev/dsp, programs such as mplayer and the experimental applications I have written will be dead in the water which is why I am asking these questions now. The experimental applications I have written are sound-activated recording and sound delays. In gcc, ioctl is what one uses to open /dev/dsp with specific characteristics such as sample rate and size. I hope what replaces it isn't too indirected. There are times when it is nice to explain what is happening in one breath. I know that isn't always possible but it is nice when you can. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150716113604.4283922...@server1.shellworld.net
Re: /dev/dsp Obsolete or Not?
Good answers. Thanks. Martin McCormick Nicolas George writes: In short: ALSA. In long: the kernel devices for ALSA are present in /dev/snd/, but applications are not supposed to access them directly, they are supposed to solely rely on the API exposed by the ALSA library, libasound. This library uses that to offer features that we would not like to be implemented in kernel space, including virtual sound devices that do not map to any kernel device. The most useful features of libasound are the plug and dmix plugins: plug converts any format provided by the application to a format supported by the underlying device; dmix allows sharing the same hardware device between several applications without a server; plus, of course, user configuration. If you are using Linux, unless you are using the third-party (IIRC, partially proprietary) OSS drivers (and you would know it), then you are using ALSA and /dev/dsp is almost completely useless. Other Libre operating systems still use the OSS API. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150716155511.46cdd22...@server1.shellworld.net
wheezy to squeeze should be wheezy to jessie
That should have been a subject of wheezy to jessie. I goofed -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150716161308.01b2022...@server1.shellworld.net
wheezy to squeeze
It is time to finish the upgrade from squeeze to jessie, I think. It looks like the squeeze to wheezy upgrade worked but I see a problem when trying to upgrade from wheezy to jessie. Here are the active lines in sources.list: When all entries pointed to wheezy, I did the upgrade and nothing showed up as being held back so I changed all to jessie and there's a problem all right: deb http://ftp.us.debian.org/debian/ jessie main deb http://security.debian.org/ jessie/updates main deb http://ftp.us.debian.org/debian/ jessie contrib non-free I also had the line: deb http://ftp.us.debian.org/debian/ jessie main contrib non-free but that triggered a warning of duplicate sources so I commented it out. 622 upgraded, 0 newly installed, 0 to remove and 551 not upgraded. Need to get 222 MB of archives. After this operation, 48.4 MB of additional disk space will be used. Do you want to continue [Y/n]? Needless to say, I typed n and there's where things stand now. The number of held-back entries is about the size of the whole distribution so something is seriously wrong. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150716160945.1c57722...@server1.shellworld.net
Re: the State of Linux Audio
The audio FAQ on the debian wiki does say that sometimes support for certain sound cards is removed from new kernels due to licensing issues. It is always possible that this is what happened but since there is a module right there in the only 3.x kernal on this system, I think that it is more likely that some sort of bug is afoot. Squashing it will be lots of fun. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150716032046.3555f22...@server1.shellworld.net
Re: the State of Linux Audio
Nicolas George writes: Le septidi 27 messidor, an CCXXIII, Lisi Reisz a e'crit : How are you getting these useful error meassages if sound isn't working? Did oyu say atht you are sshing in from a working box? Usually, error messages are to be read on the screen. Martin wrote he ran commands and observed the output, we can deduce that the non-working sound cards are definitely not the only output device available on this computer. Correct. There are ways to make them talk. I cheat by having a separate computer that runs Linux with the speakup kernel so that's how I get those messages. I also setup a serial console on the system in question and use kermit to get in if the network stack is not working on the target machine. You also get most of those messages in /var/log/syslog or by running dmesg and reading that output. There is a lot of repetitious stuff there, but some times one will find a nugget of incite to solving a problem. That hasn't happened yet in this case but I plan to look some more and force myself to read the really dull stuff that happens after the plug-and-play routine finds the on-board audio device. Somewhere, some way, that information is getting ignored as one would expect an attempt to pull in the corresponding kernel module which is not happening for some reason. If I find out something that is wrong, I will surely share it with the list as this is really not good nor is it progress. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150716005310.6479b22...@server1.shellworld.net
Re: Upgrade from Squeeze to Wheezy Killed Sound on Dell Board.
Javier Barroso writes: There is a page on the wiki [1] where give you details about cs4236 devices on Debian (and why they were excluded from Distribution. I'm not sure if cs4236B is included. I hope it work too, I looked there and didn't see any documentation stating that the 423X series which appears to run from CS4231 through CS4239 and lots of revisions such as the CS4236B is no longer supported. I also found the following kernel module: /lib/modules/3.2.0-4-686-pae/kernel/sound/isa/cs423x/snd-cs4236.ko which makes me think it should still work. The strangest thing about all of this is that even though the plug-and-play detection software sees the CS4236, the final state of the system after it is in multi-user mode sees no sound cards at all. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150715170147.790bc22...@server1.shellworld.net
the State of Linux Audio
I'm the one who has been asking questions about getting an old Dell Dimension mother board with an on-board CS4236 sound card to work again after upgrading to wheezy. For years, I have had pulseaudio and alsa on this system and have also seen what I will describe as weirdness which makes me wonder what is really working and what is not. If you do ps ax |grep pulseaudio |grep -v grep to look for any pulseaudio processes, one does find a process for pulseaudio and it is configured for per-user sessions. You can reportedly play multiple audio sources with pulseaudio but I have never been able to play more than one source at a time with further attempts to play something resulting in a device busy error which is normally not a problem but it's obviously not coming from pulseaudio. This system also has no X-windows clients and thus is a command-line-only system but I constantly see the following message in syslog: wb5agz pulseaudio[20877]: [pulseaudio] server-lookup.c: Unable to contact D-Bus: org.freedesktop.DBus.Error.Not Supported: Unable to autolaunch a dbus-daemon without a $DISPLAY for X11 This looks pretty sick to me but it could be that the daemon is working but just can't shout out to anyone which is kind of dumb since an error message looks just as informative on a console as it does in a GUI. The only reason I put pulseaudio on here was way back when I was running lenny and had no /dev/dsp. Someone suggested installing pulseaudio. I did. /dev/dsp came back and life marched on. Generally, sound got easier and better with upgrades but the upgrade to wheezy turned back the clock and sound is broken for all practical purposes. There were actually two sound cards, the CS4236 on the mother board plus an AWE64 Gold which behaves like a SBLive board in Linux. No wave tables or other special effects but recording and playback are fine. After the wheezy upgrade, both sound cards went poof and one would never know they were there except the CS4236 shows up in dmesg as a plug-and-play card. The SBLive does a better job of hiding but manages to sometimes be able to kill the Ethernet interface probably by fighting over the same interrupt. a Soundblaster Digi which is a fancy USB card that had worked fairly well both recording and playing under squeeze now limps along with only playback of the left and right channels and absolutely nothing else. Nothing regarding sound is better on this system and many things that have worked flawlessly for over ten years such as the ability to autodetect the on-board sound card and install a /dev/dsp device are all gone. I have an ace in the hole in that I had an extra boot drive so I used dd to copy the original squeeze drive to the new soon-to-be wheezy drive. After running the upgrade and loosing all the sound, I can simply slip the squeeze drive back in and there should be music again but support for squeeze is running out soon and, as a retired worker in network operations, I know that one of the best ways to be safe on the internet is to keep your computer's OS up to date. The rifraf out there will at least have a little more trouble cracking your system if it is current than they will if it is several revisions behind and all the bad guys know how to break in. Except for the sound, everything else seems to be in order though it is, of course, hard to tell for sure until you try to do something and now you can't when you could before the upgrade. Legacy code is not necessarily bad and one would hope that new code builds on the legacy as opposed to just whacking off stuff that used to work and replacing it with something that instantly renders a whole table full of equipment useless. One expects things like that from purely commercial software but one of the neat things about Linux is that it isn't or at least wasn't quite as picky about hardware. Oh well, I am dangerously close to ranting so let's stop and see what others say. Mainly, if there is a better way to do Linux sound, I'm all ears. The silence is deafening. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150715194414.359c822...@server1.shellworld.net
Re: Fetchmail may almost be working in pop3.
Paul E Condon writes: I use msmtp, not exim, even though exim comes already installed by Debian. Msmtp has its own tiny config file which can be located at ~/.msmtprc You can put there whatever you need to satisfy you ISP and have no fear of exim mucking about with it. Of course, don't remove exim once you see msmtp working. That would break you Debian installation. Msmtp is a package in the main branch of all Debian repositories. Thank you. That is great to know. I happen to use nmh to read and send messages. The newest versions of nmh which is a very old application but extremely safe from a security standpoint allow one to modify the From: line in the message. I am upgrading the system I use for email to jessie so I can upgrade nmh. If that still fails to satisfy the authentication requirements, it is good to know of msmtp. I'm always looking for a bigger hammer, so to speak. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150714124750.9bceb22...@server1.shellworld.net
Re: Fetchmail may almost be working in pop3.
David Wright writes: A number of very good suggestions The other thing you could try is a handcrafted email, which takes about 5 minutes, by typing the following into a bash prompt: $ echo -e -n '\0marti...@suddenlink.net\0SECRET' | base64 aBase64stringIsEmitted= $ openssl s_client -starttls smtp -crlf -connect smtp.suddenlink.net:587 ehlo hostname auth plain aBase64stringIsEmitted= mail from:marti...@suddenlink.net rcpt to:mar...@shellworld.net data subject: testing 587 with ehlo hostname... --- the blank line between header and body text of message is any other info saying what parameters you used. . quit Cut and paste this at the prompts, having put in your password on the first line, and copied the resulting base64 string after auth plain. It might be worth doing all this in a script command, as there's a big block of garbage emitted when you enter the openssl line. Here's an edited example of what you I see when I do this: Ah! more good things to try. I have, on occasion, manually generated various messages in years past when tormenting spammers so this would actually be a constructive use of this sort of activity.:-) Scripts are perfect for this so one can duplicate the same conditions or fix a minor syntax error without having to re-do everything. Again, many thanks. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150714130331.2e0e922...@server1.shellworld.net
Upgrade from Squeeze to Wheezy Killed Sound on Dell Board.
The system in question is a Dell Dimension 600-MHZ Pentium from way back in 2000. The BIOS date is October 10 of 1999. The sound chip set is a CS4236 on the mother board and it's always been touchy about working. You can count on the sound dying after any significant upgrade but once you get it working, it works very well for both recording and playing. The problem seems to be that automated methods for finding sound devices frequently miss this card and that is what happened after upgrading from squeeze to wheezy. The output of aplay or arecord -l shows no sound cards present even though dmesg sees the card. [1.617732] pci_hotplug: PCI Hot Plug PCI Core version: 0.5 [1.617835] pciehp: PCI Express Hot Plug Controller Driver version: 0.4 [1.617846] acpiphp: ACPI Hot Plug PCI Controller Driver version: 0.5 [1.618473] intel_idle: does not run on family 6 model 8 [1.618553] ERST: Table is not found! [1.618562] GHES: HEST is not enabled! [1.618621] isapnp: Scanning for PnP cards... [1.719015] 01:01: card 'CS4236B' [1.719026] isapnp: 1 Plug Play card detected total That is exactly what is there right now, that one card but alsa doesn't seem to know about it any more. As a computer user who is also blind, I use the accessibility features built in to the modern Linux kernel when they work and I have a stack of various boot and rescue disks such as Talking Archlinux, Vinux4.0 and the regular old installation disks for ubuntu and wheezy which all talk on many systems but not this one. The silent failure of archlinux is interesting because there is both a recording of an actual human voice at the beginning introducing the talking archlinux distribution plus speakup which is the software speech generator one gets when switching to the rescue shell. No sound is audible on this system from the sound card and I bet now that algorithm for discovering sound cards is not finding any card once again. Older boot disks such as Vinux2.0 based on lenny from 2009 talk just fine on this system so the problem almost has to be the modern algorithms for finding sound cards during boot. This old Dell is not ready for the recycling center as it has a gigabyte of RAM and can still do lots of useful work so I hope there is a way to get audio working again. I used the Vinux2.0 CD just yesterday so as to use dd to copy one boot drive to another of the same size. It was kind of scary because the device node for the boot drive came out as /dev/hda rather than /dev/sda. /dev/sda was the good drive I was copying from so I gulped, took a deep breath and did dd if=/dev/sda of=/dev/hda and got my copy without ruining the master. I probably can't do much about the CD images that don't talk, but there are lots of Dells out there so let's hope a sound-finding strategy that really does find the cards can be developed one day. In the mean time, is there a way to get the sound back on this other wise working wheezy system? I plan to upgrade again to jessie to bring it up to date. Unless there has been work done on the sound discovery process while booting, I expect the sound card to still be dead but the death occurred between squeeze and wheezy. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150714154925.46c7622...@server1.shellworld.net
Re: Upgrade from Squeeze to Wheezy Killed Sound on Dell Board.
Dan Ritter writes: A cheap USB audio device is probably a good bet. For example, http://www.newegg.com/Product/Product.aspx?Item=N82E16812186035cm_re=usb_audio-_-12-186-035-_-Product is an $8 USB device that I can verify works with Debian and Mac OS X. That is a very good suggestion. I haven't tried that yet but I have a couple of USB sound cards that have worked well on this very system in the past. For those who are curious, I usually have another older P.C. running it's debian kernel with speakup enabled for a talking terminal and then I ssh to other systems or use ckermit to a serial console on them to do work so only one computer needs to talk. If you have a serial console, one can communicate with a unix system that is very, very sick even if it can't get on to the network right now. Now that wheezy boots up, I can go to single-user mode to make backup copies of the boot drive and since this is wheezy, the device nodes are much more sane and dd doesn't make you wonder if you are trashing your master copy instead of copying it to a safe place. Again, thanks. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150715012837.4734d22...@server1.shellworld.net
Re: Fetchmail may almost be working in pop3.
David Wright writes: I don't see what the issue is. People with different usernames send mail from this system. Correct. After looking at what I posted, it is confusing. Let's try again. Do you mean /etc/mailname? What's actually in there? wb5agz.swbell.net That should never show up on the remote end because this is the host name that is locally defined and not on any DNS A record. The user ID for me on the local system is martin. The user ID I registered on the smarthost was martin.m since martin without anything else already belongs to another customer. If I hope to send mail through smtp.suddenlink.net, it must see marti...@suddenlink.net plus the password also used to retrieve pop3 mail and the retrieval does work. Do you mean the From: line in your email header? What are you typing in (or what is your mail client putting there) and where are you observing the changed version? The mainlog file displays the error that smtp.suddenlink.net is reporting 2015-07-11 06:29:26 1ZDsyD-0001Rm-PO ** mar...@shellworld.net R=smarthost T=remote_smtp_smarthost: SMTP error from remote mail server after MAIL FROM:mar...@suddenlink.net SIZE=1586: host smtp.suddenlink.net [208.180.40.68]: 553 Authentication is required to send mail as mar...@suddenlink.net I totally agree. That should have been marti...@suddenlink.net. That is where the wheels came off. The smarthost sees mar...@suddenlink.net and what it should see in the From: line is marti...@suddenlink.net Here are all the non-comments from update-exim4.conf.conf dc_eximconfig_configtype='satellite' dc_other_hostnames='' dc_local_interfaces='127.0.0.1' dc_readhost='suddenlink.net' dc_relay_domains='' dc_minimaldns='false' dc_relay_nets='' dc_smarthost='smtp.suddenlink.net::587' CFILEMODE='644' dc_use_split_config='true' dc_hide_mailname='true' dc_mailname_in_oh='true' dc_localdelivery='mail_spool' Here are all the important parts of passwd.client except the password. smtp.suddenlink.net:marti...@smtp.suddenlink.net:SECRET I hope this is a bit more clear. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150711124349.2730f22...@server1.shellworld.net
Re: Fetchmail may almost be working in pop3.
Bob Bernstein writes: what do you put in exim's config as the name of your smarthost? dc_smarthost='smtp.suddenlink.net::587' I have figured out the first thing that is wrong but am not sure how to fix it. When registering a user ID on Suddenlink's email gateway, I had to pick a slightly different version of the user name than is on the sending system because mar...@suddenlink.net was already taken. exim4 prepends the local UID to the domain portion of mailname so the From: line is mar...@suddenlink.net. I even tried marti...@suddenlink.net and it predictably built martin@marti...@suddenlink.net which is understandable but wrong. The prepending of the local UID is autimatically done so one can enter only the fqdn of the smarthost. That's where things stand now. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150711031342.015db22...@server1.shellworld.net
Re: Fetchmail may almost be working in pop3.
The job now is to get the out-bound authentication to work to the smtp server. One should use dpkg-configure exim4-config to set exim to use a smarthost for out-bound messages and rely on fetchmail for the incoming mail. Most of this is relatively easy and straight-forward except for one small detail. The box being used to send mail to the provider's smtp host does have a FQDN but it is fake and appears nowhere in any working DNS. When I put Linux on that box, I used wb5agz as the host name and swbell.net as the domain name even though we were on DSL with out a static IP address. The only other systems aware of this name are systems on the local side of the router with wb5agz.swbell.net in their /etc/hosts files mapped to a private IP address. Nowhere else in the world does this host name mean anything. It seems that the fake name is still finding it's way in to headers that go to the smarthost and it doesn't like seeing that since it doesn't resolve. Are there flags I can send to exim4 to see what the message looks like which will probably tell me which headers are wrong? This will make it possible to go through exim4-config once again to see what I set wrong since the smtp server must not try to resolve the fake host name. All it knows about is the mailname and the password. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150710220657.bf43722...@server1.shellworld.net
Re: Fetchmail may almost be working in pop3.
Bob Bernstein writes: Is there a special reason you do not post your .fetchmailrc file? Yes. This is called a senior moment. It's when you forget to include all the relevant information for which I apologize. Here is the slightly obfuscated .fetchmailrc file. The only obscured part is the password. #./fetchmailrc for martin # This file must be chmod 0600, owner fetchmail # set polling time (10 minutes) # set daemon 600 set no bouncemail defaults: mda '/usr/bin/procmail' antispam -1 batchlimit 0 #server settings #poll suddenlink.net with protocol pop3 poll pop.suddenlink.net with proto POP3 # set username user 'martin.m' password 'SECRET' End of .fetchmailrc settings Running under --Version produces the following with password redacted: This is fetchmail release 6.3.18+GSS+NTLM+SDPS+SSL+NLS+KRB5. Copyright (C) 2002, 2003 Eric S. Raymond Copyright (C) 2004 Matthias Andree, Eric S. Raymond, Robert M. Funk, Graham Wilson Copyright (C) 2005 - 2006, 2010 Sunil Shetye Copyright (C) 2005 - 2010 Matthias Andree Fetchmail comes with ABSOLUTELY NO WARRANTY. This is free software, and you are welcome to redistribute it under certain conditions. For details, please see the file COPYING in the source or documentation directory. This product includes software developed by the OpenSSL Project for use in the OpenSSL Toolkit. (http://www.openssl.org/) Fallback MDA: (none) Linux wb5agz 2.6.32-5-686 #1 SMP Wed Feb 18 13:24:13 UTC 2015 i686 GNU/Linux Taking options from command line and /home/martin/.fetchmailrc Poll interval is 600 seconds Idfile is /home/martin/.fetchids Fetchmail will forward misaddressed multidrop messages to martin. Fetchmail will direct error mail to the postmaster. Fetchmail will treat permanent errors as temporary (keep messages). Options for retrieving from marti...@pop.suddenlink.net: True name of server is pop.suddenlink.net. This host will be queried when no host is specified. Password = SOME_PASSWORD. Protocol is POP3 (using default port). All available authentication methods will be tried. Server nonresponse timeout is 300 seconds (default). Default mailbox selected. Only new messages will be retrieved (--all off). Fetched messages will not be kept on the server (--keep off). Old messages will not be flushed before message retrieval (--flush off). Oversized messages will not be flushed before message retrieval (--limitflush off). Rewrite of server-local addresses is enabled (--norewrite off). Carriage-return stripping is enabled (stripcr on). Carriage-return forcing is disabled (forcecr off). Interpretation of Content-Transfer-Encoding is enabled (pass8bits off). MIME decoding is disabled (mimedecode off). Idle after poll is disabled (idle off). Nonempty Status lines will be kept (dropstatus off) Delivered-To lines will be kept (dropdelivered off) No received-message limit (--fetchlimit 0). Fetch message size limit is 100 (--fetchsizelimit 100). Do binary search of UIDs during 3 out of 4 polls (--fastuidl 4). No SMTP message batch limit (--batchlimit 0). No forced expunges (--expunge 0). Messages will be delivered with /usr/bin/procmail. Recognized listener spam block responses are: -1 No pre-connection command. No post-connection command. Single-drop mode: 1 local name recognized. martin No interface requirement specified. No monitor interface specified. No plugin command specified. No plugout command specified. No UIDs saved from this host. No poll trace information will be added to the Received header. Messages with bad headers will be rejected. All that does match what I am expecting based on config. Here is an actual session: Old UID list from pop.suddenlink.net: empty Scratch list of UIDs: empty fetchmail: 6.3.18 querying pop.suddenlink.net (protocol POP3) at Thu Jul 9 06:19:20 2015: poll started Trying to connect to 208.180.40.196/110...connected. fetchmail: POP3 +OK InterMail POP3 server ready. fetchmail: POP3 CAPA fetchmail: POP3 +OK Capability list follows fetchmail: POP3 TOP fetchmail: POP3 USER fetchmail: POP3 RESP_CODES fetchmail: POP3 PIPELINING fetchmail: POP3 EXPIRE NEVER fetchmail: POP3 UIDL fetchmail: POP3 IMPLEMENTATION Openwave Email vM.8.04.03.20 201-2389-100-164-20 fetchmail: POP3 150330 fetchmail: POP3 . fetchmail: pop.suddenlink.net: opportunistic upgrade to TLS failed, trying to continue. fetchmail: POP3 USER martin.m fetchmail: POP3 +OK please send PASS command fetchmail: POP3 PASS * fetchmail: POP3 -ERR invalid user name or password. fetchmail: invalid user name or password. fetchmail: Authorization failure on marti...@pop.suddenlink.net fetchmail: For help, see http://www.fetchmail.info/fetchmail-FAQ.html#R15 fetchmail: POP3 QUIT fetchmail: POP3 +OK martin.m InterMail POP3 server signing off. fetchmail: 6.3.18 querying pop.suddenlink.net (protocol POP3) at Thu Jul 9 06:19:21 2015: poll completed Merged UID list from
Re: Fetchmail may almost be working in pop3.
Lisi Reisz writes: As someone else has pointed out, it looks as though your username is wrong. Most POP3 mailhosts require the full email address, with the @domain bit. Lisi This one is no exception. Thank you!! I don't know how many times I have read and re-read the lines in that .fetchmailrc file and missed that. That was the problem. When I ran it again, it downloaded two test messages lightning fast and the log showed no further squawks. I don't know whether to feel stupid or joyful. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150709132836.0fffb22...@server1.shellworld.net
Fetchmail may almost be working in pop3.
I am trying to get a debian squeeze system to pull mail from my cable provider's pop3 server. It appears they are not doing anything really out of the ordinary but I obviously have something set wrong. Here is a short snippet from their instructions for using pop: Incoming Mail Server: pop.suddenlink.net Incoming mail port: 110 Incoming mail port (SSL): 995 end of snippet It all starts out OK: fetchmail: 6.3.18 querying pop.suddenlink.net (protocol POP3) at Wed Jul 8 21:23:25 2015: poll started Trying to connect to 208.180.40.196/110...connected. fetchmail: POP3 +OK InterMail POP3 server ready. fetchmail: POP3 CAPA fetchmail: POP3 +OK Capability list follows fetchmail: POP3 TOP fetchmail: POP3 USER fetchmail: POP3 RESP_CODES fetchmail: POP3 PIPELINING fetchmail: POP3 EXPIRE NEVER fetchmail: POP3 UIDL fetchmail: POP3 IMPLEMENTATION Openwave Email vM.8.04.03.20 201-2389-100-164-20 fetchmail: POP3 150330 fetchmail: POP3 . fetchmail: pop.suddenlink.net: opportunistic upgrade to TLS failed, trying to continue. fetchmail: POP3 USER martin.mccormick fetchmail: POP3 +OK please send PASS command fetchmail: POP3 PASS * I do have a password set in .fetchmailrc and am fairly sure it is good but where does the SSL come in? Any attempt to start a session using SSL port 995 instead of 110 only causes a hung session for about 5 minutes followed by a socket error. The system does have secure ssl and does appear to work with https links. Any Suddenlink users here who have gotten a successful .fetchmailrc to work? I did try the fetchmailconf program on a X terminal and never got it to work so I may simply be missing a command. Thanks for all constructive suggestions. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150709030326.5c5ef22...@server1.shellworld.net
Re: Frequent Network Disconnect/Reconnect
bri...@aracnet.com writes: have your tried swapping out ethernet cables ? Also, have you tried another computer on the same switch port to see if it has trouble? Have there been any changes made to your network infrastructure especially to switches your system is connected to? Does the Link light blink if you can see it? I recently retired from working in our network operations group at a university, here, and we had a case where an outside organization which uses our resources installed a bunch of their switches on our network with our permission but something was wrong with the switches and they produced literally thousands of port flaps per day. We made sure our network connection to them was solid but the last time I heard any more about the situation, they were still plagued with flapping ports. If you can, try your same system and cable on another port to see if there are any improvements. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150709032337.1c36a22...@server1.shellworld.net
Re: Mail and POP3
Stuart Longland writes: I've done this before with numerous distributions in the past. Basically you set up fetchmail to do the mail collection, and I think by default it tries to use the local delivery agents to deliver mail to local users. So you set it up as a daemon to collect mail for a number of users. Your SMTP server then looks after local delivery and for delivery to a smarthost outside your network (your ISP). I don't recall what the exact configuration parameters are for fetchmail, it's been a while since I've used it, but there is one that controls who email from a particular account gets delivered to. Once you set that, and assuming your SMTP server (exim4 in your case) is set up correctly, things should JustWork?. Thank you and others. This all makes sense. I've got about a month to make it work which is hopefully about 30 more days than I need. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150630124117.be8f722...@server1.shellworld.net
Mail and POP3
This system runs debian squeeze for now and I want to make it use our internet provider's POP3 mail server and send out-bound mail through the provider's smtp server. In the past, I have used similar systems connected to the internet so I simply configured exim4 accordingly and things worked fine. I found an example for debian-etch which used fetchmail. Is that still the case for squeeze and newer debian releases? Do I need to leave exim4 alone as it appears that fetchmail does all the moving? Thank you very much. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150630014405.59d3522...@server1.shellworld.net
Debian wheezy exim4 Refuses All External Mail.
I am more used to sendmail under FreeBSD and I suddenly lost my FreeBSD system on which I receive mail from everywhere so I need to quickly make a wheezy system stop rejecting all incoming non-local messages. The exim4 installation on the system in question is the out-of-the-box installation that came on the wheezy installation CD and every indication is that it is working as it should right now. I want to make it receive all mail and deliver it locally to users on the system which is me. There is rc.local and bogofilter on the system for spam control and sorting of messages to appropriate folders, but right now, this system and another I have access to always reject any connection other than telnetting to port 25 on the local box. What is the simplest and safest change I need to make to open the systems up to external mail? If I can get this system receiving mail normally, the FreeBSD virtual system can wait but time is getting tight and I haven't found the magic command yet. Many thanks. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20141107123121.0b8d622...@server1.shellworld.net
Re: Debian wheezy exim4 Refuses All External Mail.
Joe writes: much good information not quoted but greatly appreciated etc. and try to telnet in from outside, see what message you get. 2dc martin tmp $telnet debsystem.it.okstate.edu 25 Trying 169.254.5.10... telnet: connect to address 169.254.5.10: Connection refused telnet: Unable to connect to remote host 3dc martin tmp $ Just kidding about the IP numbers, but that's how the session went. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20141107141006.1525322...@server1.shellworld.net
Re: Debian wheezy exim4 Refuses All External Mail.
Joe writes: original state. Either way, check /etc/exim4/update-exim4.conf.conf, which gets updated by dpkg-reconfigure. The file contains instructions as to how to make changes. This has gotten me started on the right direction plus, of course, man update-exim4.conf. The important clause, here is .ifndef DC_eximconfig_configtype DC_eximconfig_configtype=internet .endif This is as opposed to local One other thing is mentioned in the man page that I must do. I run our domain name server and will create an MX record for the dead FreeBSD VM (RIP) so that mail that would have gone to it will go to the Debian box. When you do that, one must make the mailer on the target system aware that it is supposed to receive mail addressed to the system in the MX record. There is dc_other_hostnames for this purpose but it is not a configuration variable as such. If you run update-exim4.conf -oexim4.conf.conf, you don't see the syntax for how to enter this list since it wasn't there to begin with. All the variables are ifthen clauses. Do you just tack this list on at the bottom like dc_other_hostnames deadmouse.okstate.edu or is it more structured? I think that's about all I don't understand for now. In the 13 or 14 years I have been using Linux, this is the deapest I have gotten in to exim. It is certainly different from sendmail but so far, I like it. Anyway, thanks for the help. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20141107153419.3c25422...@server1.shellworld.net
Debian wheezy exim4 Refuses All External Mail. Solved
Joe writes: You're in the wrong place. First, exim4 can use either one large main configuration file, or it can use many files for individual configuration options, and you were asked to decide which in the original configuration questionnaire. In this case, it doesn't matter which you chose. If you chose one file, it will be exim4.conf.template, otherwise it's the files under conf.d. Either way, that's where you find all the if/then clauses, which are used to merge update-exim4.conf.conf variables into the main file(s). The file you want is update-exim4.conf.conf, and I copied mine almost verbatim. There are no if clauses there. Each line just begins dc_xxx and ends with ='yyy'. Just edit between the quotation marks, save, and run update-exim4.conf. Domain names go in the 'dc_other_hostnames' and if there is more than one, they are separated by colons, with no spaces. Check after the update that /etc/exim4/update-exim4.conf.conf still looks right, similar to mine. Restart exim4 after that, I don't believe that running update-exim4.conf does that. The main configuration file, exim4.conf.template, or the individual files, allow all possible configurations, but in an effort to keep the local user adjustments out of these files, the dozen or so most fundamental configurations can be made easily either by editing update-exim4.conf.conf or answering the questions asked by dpkg-reconfigure exim4-config. If you need to do more complicated things, you do need to pick your way between ifdefs in the main configuration file(s), but that isn't necessary just to get the basic operation right. I ended up doing the dpkg-reconfigure exim4-config command mainly just to see what questions it asks and it is not bad at all. If you want to go through it again, the previous choices become the defaults so one can just hit Enter for most of the questions. Your example was good to compare with but I figured that I needed to go through the whole configuration since what you start with after installing wheezy does not include /etc/exim4.conf.conf. I'm now getting mail from all over and need to get procmail working on this system to bring some sanity to the torrent of messages coming in, a lot of which are from the mailer daemons on systems trying to deliver mail over the last couple of days. Thanks for all the help. I will save the messages for future reference. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20141108022649.5adc822...@server1.shellworld.net
Re: Slight New Sound Problem
T.J. Duchene writes: Martin, I'm sorry you had problems with my suggestion. Most often, these problems have to be handled by trial and error. I'm afraid I can only offer advice based on my own experience and the fact you mentioned you were using Pulseaudio. I assumed you had it already installed and was using it. So did I. If you hadn't gotten me checking in to pulseaudio, I'd still think it was actually doing something useful. Thanks for the assistance as it is just as important to know what isn't a factor as it is to know what is. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140920113929.a842622...@server1.shellworld.net
Re: Slight New Sound Problem
Chris Bannister writes: I reckon the guys on the 'linux-audio-user' (http://lists.linuxaudio.org/listinfo/linux-audio-user) mailing list would be the ideal place for help with this. Probably so. I've exhausted all the obvious solutions now. Thank you. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140920114632.8f9eb22...@server1.shellworld.net
Re: Slight New Sound Problem
T.J. Duchene writes: Pulseaudio has had a long history of being poorly handling certain audio chipset drivers, I'm afraid. You may be able to solve your problem by adjusting the the driver parameters in the file: /etc/pulse/default.pa. The more I dig in to this, the less I know. Back in 2009 or so, I upgraded the system in question to squeeze and I remember discovering that /dev/dsp had gone away so I was advised to install pulseaudio. I've slept a few nights since May of 2009 so I do not remember everything that happened but I imagine I did apt-get install pulseaudio. I did get /dev/dsp back and have successfully used it ever since. I now have two sound cards so there is /dev/dsp and /dev/dsp1 and they mostly work except for the glitches. Alsa is also here and I think that is what is really working. There was no /etc/pulse/default.pa and no executable file named pulseaudio anywhere on the system as well as no man page for pulseaudio. So, I decided to purge pulseaudio and start over as there was a libpulseaudio on the system and /etc/pulse/client.conf. apt-get reported that pulseaudio was not installed so I couldn't purge it so I did apt-get install pulseaudio and got it. This brought more files in to /etc/pulse including the default.pa file and the executable for pulseaudio and it's man page. pulseaudio can be started and runs and has no effect on the glitches nor does it break anything. It's just another running process. It appears that pulseaudio can be an audio server and do many things but I am not sure how it is supposed to fit in to the scheme of things. There is no native .pulse directory in my home directory and, when I made one, nothing appeared there even after I briefly ran pulseaudio. This system goes back to the lenny distribution and it's sound was, at one time, 100% derived from alsa. /dev/dsp always blocks if you feed it more than one source so I am wondering if there is some old legacy configuration file gumming up the works. Sound basically works on this system with the exception of the random glitches. If not for those, I would still think that pulseaudio was running when it really wasn't here. Like various famous people of our past have reportedly said, It ain't what you don't know that can hurt you but what you do know that just ain't so. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140919113013.c6a9222...@server1.shellworld.net
Re: Slight New Sound Problem
Marko Randjelovic writes: Did you try with another kernel? Well, indirectly. As I mentioned, the system has always exhibited this behavior slightly for several years through a number of kernels. The biggest change, though, was when I changed out the conventional 10 GB hard drive for a slightly larger flash drive that was also about 15 years newer. I think it is some sort of bus contention problem. The system has two IDE controllers. One has the boot drive on the master position plus a second conventional hard drive on the slave position for /home. The other IDE controller has a CDRW drive in the master position and a second CDRW drive in slave. I can always make the sound problem worse by doing disk-intensive activity on the controller that has the two fixed disk drives. The system kernel changed in May and there was no noticeable change then that could be tied to the new kernal. Thanks for the good question. It made me think differently. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140918111400.d863b22...@server1.shellworld.net
Re: Slight New Sound Problem
T.J. Duchene writes: Good morning, Martin! Before I can make suggestions, I need to know if you are using a daemon such as Jack or PulseAudio or if you are using ALSA directly. Thanks, I am using pulseaudio and alsa. Normally, if I am listening to something it is through mplayer but aplay also is effected by the problem. It seems that any high-quality audio application now is showing the glitches. As an example, I wrote a C program a few years ago that turns the sound card in to a variable-length audio delay. When OSU has a football game on both TV and radio, we want to hear our home sports announcers and see the game on TV. Usually, the radio is 10 to 20 seconds ahead of the video and my trusty delay makes it possible to get them both synced. The card is set to a 32000 sample-per-second rate and /dev/dsp is opened for writing at the start of the program. The read pointer is set to a character in the buffer that is far enough away to equal the needed delay. the write and read pointers chase each other round and round the buffer. I can now hear the glitches on that application, also. A hint to the wise, if you write a delay like this you had better write half-level silence values to the ring buffer when initializing the program or you will hear seconds of extremely loud static thundering out of the speakers until the read pointer finally sees output from the sound card. With the initialized buffer, you hear nothing until sound comes out. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140918155811.a3e2522...@server1.shellworld.net
Slight New Sound Problem
This is an older Dell system whose on-board sound chip is a CS4237 and it has worked well until I replaced the boot drive with a flash drive. This makes the system faster but audio now has a problem that I would sure like to correct as it is annoying to say the least. I began noticing it when listening to mixtures of voice and music so I played a steady 400-HZ tone for several seconds and listened to it. Every 2 to 5 seconds, the tone takes a very small hit. Sometimes, the hit is in the form of a small tick as if the samples sped up and other times, the tone takes a hit that sounds like the samples slowed for a tiny fraction of a second. I thought it might be related to kjournald writing to the flash drive so I found a command one can place in to /etc/sysctl.conf that sets the write schedule for updating the journals. This time, I set it to a minute to see if anything changed. No change at all and the hits just kept on coming. I also temporarily took out the commands in /etc/modprobe.d/alsa-base.conf that determine the order of the sound cards and that had no effect. This system actually has always had a tendency to do this very slightly but the glitches got worse after the new hard drive. Except for that, the CS4237 has always made good recordings and playback. The little hits are now big enough, however, to be audible during normal speech. Another thing I tried was to change the nice number of either mplayer or aplayer while it was playing the tone to see if the problem got better or worse. No change at all. It sounds as if it might be a buffer issue since, as I understand it, the sampling rate is clocked in the card and the buffer should have enough capacity to not run out as long as playing or recording is being done. Before I changed the boot drive, I could sometimes damage a recording by lots of file activity on the second IDE drive on the same controller as the boot drive. The glitches are not audible at low sample speeds such as 8000 Hertz such as communications-grade audio but are most pronounced during 44.1 KHZ sampling. Any ideas on anything else to tweak? Thanks for any constructive ideas. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140918023227.99a8722...@server1.shellworld.net
the Mysteries of asound.conf
If one searches for debian+multiple+sound+cards, there
is a wilderness of somewhat confusing discussions and examples
as to how to configure asound.conf to insure that each card
comes up in the same order. I have two older Dells, each with
the stock CS423X sound chip on the mother board and a SB16-type
sound card. One uses an EMU10k1 driver and the other uses an
EMU8000. All seem to work and would work a lot better if they
always came up in the same order.
One school of thought tells us to put a line in
/etc/asound.conf that looks something like
options snd slots=,snd-Emu8000
The , is supposed to cause the Emu8000 (SB) card to
always b C1 so that the CS423X always becomes c0. I then
discovered that I did not have ecasound installed which appears
to be what gives you asound so I installed it and immediately
got that long message which is the error output of the parser
that translates in to a sort of check-engine light stating This
is broken. You figure it out.
After searching for the error, I found what appears to
be a newer way to set things in asound.conf:
pcm.!default {
type hw
card 0
}
ctl.!default {
type hw
card 0
}
The syntax gods love this and aplay -l reports the cards
in the desired order but they were in the desired order to start
with so I am wondering if this actually does anything.
When booting the system from a cold boot, the CS423X
usually comes up as card 0 and the SB is Card 1.
A warm boot or a cold boot during the New Moon will flip the
order and the SB is on bottom, so to speak.
If that asound.conf example does somehow force the same
order each time, I am fine with that but it intuitively looks
like that whatever rings in first is C0 today.
Since the two sound cards are different in every way but
their function, anything that differentiates one from the other
should cause a predictable result every time.
Thanks.
Martin McCormick
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive:
https://lists.debian.org/20140829133859.060aa22...@server1.shellworld.net
Re: the Mysteries of asound.conf
Raffaele Morelli writes: drop a custom module config in /etc/modprobe.d/ eg. /etc/modprobe.d/alsa.conf and use options/index parameters That worked like a charm as far as I can tell. Thanks to both posters. I actually used the wrong module name for Card 1 and what happened was that the system came up after a boot only showing Card 0 and no Card 1 at all from aplay -l. I looked at lsmod again and realized the mistake, changing the module name for the SB card from snd_sb16 or something like that to sbawe. On the next boot, it came right up and both cards now show in the correct order. Both of these reboots were warm reboots and the order would have normally reversed by now. I think this is probably fixed now and I think I have learned a lot. Thanks. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140829155344.c072122...@server1.shellworld.net
Re: The Fine Art of Making a Bootable Drive; more
Stefan Monnier writes: One last step may be necessary : update the UUIDs in /etc/fstab and /boot/grub/grub.cfg, as you created new volumes with new UUIDs instead of cloning them. Or alternatively, change the UUIDs on the new disk with tune2fs, mkswap... to match the ones on the old disk. Otherwise you'll be stuck at grub's menu. Or just say no to UUIDs ;-) I was ultimately successful. Here is what I did: I started out with dd if=olddrive of=newdrive and waited that process out. Then, I ran fdisk -c=dos since the new disk had the older dos-compatible format thrown in for free by the dd operation. I nervously deleted all but Partition 1 and then used tune2fs to widen Partition 1 to near 15 GB out of 16 GB possible. Finally, I made another primary partition #2 to cover all remaining space and then over-wrote that with an extended logical partition 5 and made it type 82 for swap. It works like a fine watch. I finally had one last 16 GB flash drive for another old system so I simply used DD to clone the drive I had finish making on to the new drive. Of course, one needs to be careful to mount the new drive once and change such things as the host name and or any hard-coded host information such as the name and network settings or you will have real trouble if the two twins are on at the same time. Since these are flash drives, I compromised between having fstab mount / normally and mounting / with the noatime option. The relatime option is said to be less wear on the disk and still usually gives the kind of time stamping that the normal mount gives you. Thanks, everyone, for your helpful suggestions. I will save the messages for other situations in which I may need to do similar things. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140828170436.35e5722...@server1.shellworld.net
The Fine Art of Making a Bootable Drive; more
I am the one who posted stating that I can't seem to make a bootable new hard drive for my Linux Squeeze system. It's been quoted, It ain't what you don't know that will hurt you, but what you know that just ain't so. I think I am in that territory now. What I have been doing was to format the new active partition (sda1) with ext4 or ext3, using the rest of the drive space as extended primary Partition 2 and overwriting with a logical Partition 5 for swap. I then would use rsync to copy all of the old drive including special files to the new drive and one could see /dev and all hard links to initrd.gz where they should be. The final step which seems to be the kiss of death is to use grub-install on the rescue disk to put a MBR on /dev/sda. I never got this to work. Today, I did dd if=olddrive of=newdrive which worked but produces a logical drive exactly the same size as the old drive which is 10 GB. Since the new drive is 16 GB, that wastes 6 GB of capacity which is why that is not ultimately acceptable. Just for fun, though, I tried the truncated new drive and the system booted right up which proves that it is not hardware. It is the way I am building the drive. I can probably use tune2fs to re-size this new drive by blowing away the extended swap partition, moving the upper boundary of Partition 1 to 15 GB and then making a smaller extended Partition 2 with swap overwrite, but I am curious as to why the first method simply has never booted? The bad drives I created for Debian Squeeze were also formatted with ext4 for one attempt and ext2 for a different attempt. This is because the new drive is a flash drive and will need to be mounted to take wear and tear of multiple writes in to consideration. The old drive was formatted ext3 so the truncated new drive is presently also formatted the same way. Now that I know the hardware is not causing the issue, I just need to understand why or at least get some theories as to why what seems like a logical way to build a drive doesn't work. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140813130941.3426922...@server1.shellworld.net
Re: The Fine Art of Making a Bootable Drive; more
AW writes: 1. As far as I know, it's not possible to simply copy a working /dev tree. These are special files which are generated with the mknod utility. 2. Booting a computer is fairly complex. Everything needs to be at a specific location on the drive, needs to occupy the appropriate sectors - which vary in precise size depending on the drive geometry as well as the partitioning. And everything needs to appropriately connected together. 3. dd copies at the bit level. It's a low level utility. And that's why it works, while the high level rsync or cp utility will not. This certainly makes sense to me but it has some rather interesting disaster recovery implications. In this case, I am just going to a newer and slightly larger boot drive and I am lucky to have both the actual hard drive and a thumb drive copy of that drive to experiment with. The thumb drive copy is also a dd clone of the original hard drive and is obviously good because it was what I used to make the new boot drive. If one was having a bad day and their boot drive made a horrible noise and blew out a cloud of aluminum and iron oxide dust as the consequence of the meeting of a read/write head and the surface of a platter, they have no options save for recycling of the materials in the old drive. If they want to restore their old system, they must be able to restore the boot drive before applying their backup media whatever that happens to be. Chances are very good that the new boot drive will be larger or different in some way from the old one. I am not disagreeing with what you said, but it sounds like it could be a lot of trouble to restore that system. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140813152753.135fe22...@server1.shellworld.net
Re: The Fine Art of Making a Bootable Drive; more
Bob Weber writes: I use sysrescuecd (http://www.sysresccd.org/) to make a new drive bootable. There are two ways to get a bootable disk with sysrescuecd. One way is to use a special boot mode where sysrescue starts its own kernel to a system on the hard disk. Once booted you can just use 'grub-install /dev/sda' to install grub on the boot drive. I run software raid1 so I do this for both drives just in case I need to boot from sdb. A second way is to start sysrescuecd normally and mount the root file system to a directory. Make a directory say x and mount the root filesystem on it. Run these three commands: mount --bind /dev x/dev and mount --bind /proc x/proc and mount --bind /sys x/sys. Then run chroot x /bin/bash to get a command prompt running off of your root file system with the dev, proc and sys populated correctly. Now you can run the grub install command and hopefully get a bootable drive. The first method works the best since sometimes grub gets confused in the chroot environment and cant find the hard drive you want to install it on. Ben there but I wasn't sure about in which order to do the mount commands you illustrated so now I think I understand enough to get something working. My thanks also to Gary Dale in the previous posting. I will save all these messages since I haven't finished with the new drive yet but I have a better understanding of what is going on and why I was having so much trouble. Thanks to all. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140813163606.758d222...@server1.shellworld.net
Re: Mounting a FreeBSD USB Memory Stick Image rw
Zenaan Harkness writes: Martin, it looks like you'll have to recompile your kernel first sorry. I was kind of thinking that. Actually, I think I have a solution which I hadn't thought of at the time. I have FreeBSD running in a virtual machine on a Mac. That will be native ufs and I should be able to work on it there and then copy it back to put on the thumb drive when I am done. Well, at least I know for sure ufs and the Linux kernel don't just work together out of the box. I am not really complaining but I wasn't sure if they might work and I was just missing something. Again, thanks to all. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140811141443.7e8fb22...@server1.shellworld.net
Mounting a FreeBSD USB Memory Stick Image rw
Is it possible to mount the FreeBSD USB iso image on a debian system? I need to edit one of the configuration files and the nearest USB port is on a Debian system. The hope is to add a line of text to a file, transfer the image to a USB drive and boot the FreeBSD system from the memory stick for an installation. I did try mount -t ufs -ro loop FreeBSD-9.1-RELEASE-amd64-memstick-headless.img /mnt first to see what would happen and it appeared to work but ls /mnt throws an I/O error as does any operation on /mnt until one umounts /mnt. Thank you. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140808155236.6aee322...@server1.shellworld.net
Re: The Fine Art of Making a Bootable Drive
It turns out that the reason I never thought of using mkfs to build a working boot sector is that mkfs doesn't do that. Grub, however, does but I am still a bit confused as to how to get it working. I mounted the new drive on /mnt #mount /dev/sdf1 /mnt It's all there. #chroot /mnt / is now the top of that directory. #grub-install /dev/sda /dev/sda isthere but grub reports it can't find that device and is /dev mounted? This is one of those time you wish siri was part of Linux and you could yell at her! Seriously, how do I safely do this so that I make this disk think it is /dev/sda which will be what it is when it really boots? Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140804113335.d701c22...@server1.shellworld.net
Re: The Fine Art of Making a Bootable Drive
Tom H writes:
Are you mounting /mnt/{dev,proc,sys} before chrooting?
No. I did try the mount command after chrooting which successfully ran, but
didn't fix the missing /dev. I bet this is the crux of the
problem, however. Mount just mounts everything in /etc/fstab. I
don't remember if dev is there but /proc is there for sure
When I mount /dev/sdf1 on /mnt and do a ls -l /dev/sda,
it looks good.
brw-rw 1 root disk 8, 0 Jul 28 19:09 sda
Do I need to mount those befor chrooting?
The only thing I am concerned about is of course is not
overwriting the good boot sector on the old drive.:-(
Thanks.
Martin
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive:
https://lists.debian.org/20140804145710.8071922...@server1.shellworld.net
The Fine Art of Making a Bootable Drive
I thought I had a pretty good idea how to do this but I obviously am missing something. I am replacing a nearly 20-year-old 10 GB conventional hard drive with a slightly-larger flash drive for / on a Debian-squeeze system; / on flash as it were. I know this can work as I have an older version of debian on another box that has been doing this now for a couple of years and running just fine. On that system, I used dd to copy everything including the boot sector from that 10-GB drive to the new 16-GB flash drive. At that point, I had a 10-GB flash copy of every byte that had been on the electromechanical drive. I then resized the #1 partition to take advantage of the larger new disk and it ultimately worked but this can be done without quite so many steps. I used fdisk to format a brand new 16 GB flash drive such that Partition 1 is a bit over 14 GB and the rest is Logical Partition 5 and called swap. Partition 1 is marked as bootable but, at the time, I did nothing about a boot sector. I then used rsync and told it to copy devices which it appears to have done. It copied devices, /proc and /sys and I ended up with the new drive looking just like the old one except for being 6 GB larger. For the boot sector, I copied the first 446 bytes of the boot sector on the old drive as in the following example I lifted from a Google search if the two drives are different capacities: Copy MBR only of a hard drive: dd if=/dev/hda of=/dev/hdb bs=446 count=1 The last 64 bits of the 512 mbr contain partition information and this is where I may be all wet. I thought the disk-copy process took care of that but if not, this is why my new disk just sits there when it is installed. The old disk boots with no problem. There appear to be no hardware issues involved, here. The new drive is a SATA flash drive connected to an IDE to SATA converter. The little master jumper is set right and as I said, another system uses an identical hardware setup with no issues. Finally, this particular Dell mother board gives you two high-pitched beeps any time it is unhappy about hardware. It gives the beeps if the master drive is not set to be the master or is missing. In this case, it gives no beeps but also never boots. Do I need to set the top 64 bits of the boot sector? If so, how? Thank you. Martin WB5AGZ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140803124554.d76a122...@server1.shellworld.net
Re: The Fine Art of Making a Bootable Drive
I knew there would be several suggestions for solutions to making a new boot disk and I appreciate all of them. I also appreciate the explanation as to why my previous attempts at creating a bootable copy failed. It all makes perfect sense now. I will probably try mkfs first. I have used mkfs lots of times to lay down a new file system on a new disk or when re-building an older one but I never even thought of mkfs in this context so thanks for reminding me. This gives me a lot of things to try and I am sure I will have a working boot disk in the next day or so. Thanks again to all. Martin McCormick WB5AGZ Amateur radio is the original electronic social network. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140803234123.218d122...@server1.shellworld.net
A Regular Expression Question
I am trying to remove the or ampersand sign from some perl code I wrote as it is not necessary. I have no trouble finding all the instances because they consist of an ampersand immediately followed by a letter so [a-z] describes the case perfectly. The replacement pattern should actually be the same as the search pattern except that it is missing the or ampersand. The scripts have loads of logical 's and 's in them so the regular expression seems to be the safest way to globally replace everything with just everything. In this case, a global pattern that returned the search pattern missing it's first letter would even work. Thank you. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140610190120.f358e22...@server1.shellworld.net
Re: A Regular Expression Question
Reco writes: Try it like this: sed -r 's/([a-z])/\1/g' your_perl_files It worked like a charm. I forgot about the parentheses and the \1 to limit the number of matches. My thanks to everyone who emailed me both on and off-list. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140610204246.f164922...@server1.shellworld.net
I need to hear PC Speaker Beeps through the Audio Output.
Is there any kind of debian module that will reproduce PC speaker sounds through the sound card? The system in question has a working sound card but there are no apparent pins on the mother board that carry the timer-counter output to the outside world. There is a piezo transducer on the board that beeps just fine but there are no pins short of unsoldering the piezo speaker and putting a couple of pins in it's place and that is way too much work and the system does not belong to me. I need to hear the beeps in headphones. The sound card even has a 2-pin input marked PC Speaker but getting the signal there is the issue which is why I am asking if there is a software-based solution. Thanks for any ideas. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140512170029.b679722...@server1.shellworld.net
Re: Should hostname always be lower case?
The domain name system is totally case-insensitive so FiReFlY.HardknocKs.cOm will lookup as firefly.hardknocks.com. I have been administering the domain name servers for the okstate.edu domain since around 1992. We use dynamic DHCP name registration and the names that folks register are loaded with mixed-case names or names that are all upper case. The only time it matters is when we are trouble-shooting a problem and someone is looking for a certain name in a zone transfer and the name they are looking for is stored with upper or mixed case. The person working on the zone transfer must always remember to do something like a grep -i so that they don't miss the right name but in a different case than was expected. Martin McCormick Balint Szigeti writes: most of. if you will use spacewalk and your hostname will contain capital letters, it will cause problem On Mon, 2014-05-12 at 15:03 +0100, Ron Leach wrote: List, May I ask whether the 'case' of the hostname is important, in the sense that use of any upper-case characters may disrupt name resolvers? We have a server running wheezy, and we named it D7Server (not d7server). Could this matter to any resolver attempting to look up D7Server? (It might matter, for example, if standards allow resolvers, generally, to assume that names to be resolved will be lower case.) regards, Ron -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140512181046.5d12f22...@server1.shellworld.net
Re: I need to hear PC Speaker Beeps through the Audio Output.
Jerry Stuckle writes: I'm not aware of a module - but there are a lot I don't know about, anyway. Just one note - it's the transducer which actually generates the sound; it's being fed with a DC voltage. Connecting headphones to this will only give you a click in the headphones (and probably blow the board due to the low impedance of the headphones). There are piezo buzzer that actually work that way such as the Sonalert modules found in smoke alarms and zillions of other electronic devices that need to beep but the PC speaker gets it's signal from the output of a thing called a timer-counter chip whose part number escapes me at the moment. The chip was used in the original IBM PC's and consisted of a pair of timer-counters. One was set to divide the system clock speed down to 18.2 Hertz for the real-time clock and the other half could be programmed by the user to set various devisor values for the counter which then produced a square wave output that drove a speaker or piezo transducer. The old PC's also had a transistor between the timer-counter output and the speaker because CMOS chips don't have the wattage to directly drive even a small speaker to any reasonable level. I used to do a lot of 8086 assembler programming and played a lot with that speaker driver. You could actually turn the output of the transistor on without starting the timer and hear a single click or turn it on with a count stuffed in to the counter and it would generate a tone depending on what frequency you set in the counter. IOW, don't even THINK of doing it :) I probably won't go that route but one can do things like that if one is careful and makes sure not to load the output more than it was meant to be loaded or you certainly will trash the board or cause other unintended consequences such as sporadic glitches in it's operation. Again, thanks. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140512184925.00ec722...@server1.shellworld.net
Re: I need to hear PC Speaker Beeps through the Audio Output.
I performed lsmod on the system in question and it does show that lsmod loads but the second column is headed as used by and there is a 0 there which sounds like nothing is using it. Is there something I need to do to get it producing sound over the audio output jack? As I previously described, I get perfectly good stereo sound via alsa just no terminal beeps. I have the 'beep' utility installed which also makes use of the PC speaker not the sound card. If I tell beep to emit 400 for 10 seconds by beep -f400 -l1, it comes out that piezo buzzer for 10 seconds. Absolutely nothing comes out through the sound card that sounds like a 400 HZ square wave tone. The amixer settings for this particular sound card show Beep turned on and set for full volume. I am not sure if that means the hardware input on the sound card or audio input from a driver. Severl older Dell systems I have will relay the PC speaker's audio if you tell amixer to open the appropriate audio channel but they also have physical connections to the mother board's speaker driver which comes from the BIOS post test. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140513004318.a218622...@server1.shellworld.net
Re: BST Solution Worked Fine.
Lisi Reisz writes: For the archives: Note, BST was the correct *result*, *not* the *solution*. The solution was to chose /Europe/London time which will correctly switch from GMT to BST and back again as appropriate. Correct. Also for the record, when I set my system time to Europe/London/ (my hardware clock is set to UTC), my system, for some obscure reason, determinedly changes it to /Europe/Guernsey. As this is the same, it doesn't matter, but some day I might try to find out why it does it. ;-) Most likely, there is a hard link involved somewhere. London is a place the whole world knows and Guernsey may be a little outpost of Heaven, for all I know, but I bet somebody, somewhere did some customization, possibly when building the distribution image since the idea is to pick a name that most people recognize. You could do the same thing here with America/Stillwater. The Stillwater I live in is in North-Central Oklahoma and contains Oklahoma State University as it's biggest industry. There is also Stillwater, Minnesota about a thousand miles to our North. It also uses Central time and it's main industry, from what I hear is the Minnesota State Prison, a necessary but slightly different type of institution. It's residents also learn a lot there and 90% of them will eventually go forth in to the world and try their new skills so if you used /America/Stillwater for Central time, you get at least two for the price of one. Yes, a bit of a joke but check your callendar. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140401113149.8fba422...@server1.shellworld.net
BST Solution Worked Fine.
I was the one who wanted to record a couple of radio shows from the BBC and not have to remember to juggle chrontab files for the several weeks when the US is using DST and the UK isn't. This occurs in the last week of October each autumn and for around 4 weeks in March. I have an old Del Optiplex that is mainly used as a terminal to get in to other unix systems so I set that system to /Europe/London time rules and then set chron jobs based on British time. These chron jobs are just ssh commands to another old Del Dimension that has more disk space and memory to actually do the recording. Since last weekend was the UK's time change, I checked and was pleased to see that everything worked. There are more elegant ways to do this, of course. If one has a monster installation with multiple gigabytes of RAM and suitable processor capabilities, one could run a bunch of small virtual machines, just enough for a Linux installation and a few megs of space for chron and have those virtual systems set to different time zones as you needed them. If you look this topic up in Google, you do see various clever things people have tried to have part of their system pretend to be somewhere else, but that /etc/localtime rule set truly rules. One idea I thought of would be an extra field in chrontab specifically for time zone information but that would certainly add to chron's complexity so I am certainly not complaining. I have been using unix of one flavor or other since 1989 or 1990 and think it is one of the best tool kits ever. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140331201932.c0e0f22...@server1.shellworld.net
Re: Time Zone Questions
Jerry Stuckle writes: That wouldn't work well. Remember, computers are not the only ones which use UTC - in fact they are the most imprecise. There are many clocks around the world which are synchronized with UTC via radio, i.e. WWV/WWVH in the United States, CHU in Canada, and other stations around the world. I once accidentally got a Linux system set up with the wrong algorithm which I don't really remember what I did, but it maybe added the leap seconds twice as it was twenty-something seconds slow which is not a good thing at all. The hardware clock, being nothing but a crystal oscillator tied to a counter whose final resolution is 32 bits can be extremely accurate but unless you are running some sort of specialized setup, you can expect some jitter due to all the tasks your system is doing at any given time. I have an old Dell Dimension with a 600-MHZ processor and use NTP to keep the time synchronized as well as possible, but I notice that chron jobs seem to have a window of about a second in which they can fire. Sometimes, it is right on the dot and the next time it may be a second late but not much more. By the way, those Atomic wrist watches and clocks you can buy listen to something called WWVB which transmits at 60 kilohertz from the same location as WWV. There is no voice or tone for a human to hear, but the signal transmits a 59-bit word if one wants to use that term, which contains several BCD digits expressing what the UTC year, month, date, hour and minute will be at the next 0TH second. It's a leap of faith that civilization will still be here then.:-) There is at least one extra bit that sets during DST and clears in Winter. 60 KHZ is called ELF or Extremely Low Frequency and hugs the Earth. Our WWVB has been received in New Zealand though this is not a trivial task. Martin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140322135805.69b5722...@server1.shellworld.net
Re: Time Zone Questions
On a properly-working unix system, the hardware clock is set to UTC. In theory, every unix system in the world has a hardware clock that reads the same value at the same time. The localtime file is a set of rules that adjusts your UTC clock value to whatever local wall clock time should be. In a lot of Europe, it will be UTC+1 plus whatever the rules are for where you live which is why the city names exist. I live about 800 miles or over a thousand kilometers from Chicago, but that is the city those of us in the US-Central time zone set up as the localtime file because Chicago keeps exactly the same time as everybody else in this time zone. In the UK, you could just set things up for UTC, but you need the rules file Europe/London to automatically set your clock forward an hour on the last Sunday in March which is March 30 this year. In some parts of the world such as the Northern Territory of Australia and at least parts of India, the time correction is designed to be closer to Solar time so while the hours all change at the same time for most of us, their hours change on our half-hour. We had a student working for us a few years ago who was from India and he told me that his home was 9-and-1-half hours ahead of Central time. I don't have any idea if this value is constant all year but it most likely varies when each country adjusts it's clocks for daylight shifting, whatever you like to call it. Still, if you dug in to the computer of a resident of India or the Northern Territory of Australia, red their hardware clock and then immediately read the hardware clock of a resident of London or Las Angeles, they would, in theory, read exactly the same count. Of course, if you have a computer that makes use of two operating systems such as Windows and Linux, you may have to forego all that great automation and set your hardware clock to local time and remember to reset it when the clocks change. Actually, I think Windows now also uses the UTC plus local rules method of keeping it's time. Anyway, I have some old Linux systems which are all using America/Chicago except for 1 which is using posix/London which will hopefully make cron run as if I were, in fact, in the UK. To answer Ron's question, the time stamps on files are based on the hardware clock to the best of my knowledge. When you look at one, a binary value reflecting what the hardware clock was is neatly converted to the text you see so, if the rules change, your older files might appear to have been made an hour sooner or later than they really were made unless the rules file remembers when the rules changed and adjusts for that. I bet you never thought it was this complicated. I could be wrong about it all, but I think most of this is accurate. Martin Ron Leach writes: Hadn't realised any of this, so thank you. If 'system time' and 'desktop time' differ - such as is suggested - what 'timestamp' is put on files when they are created? And does this differ whether the files are on NFS, and on another server? Is there an implication, here, that if a site uses desktop and system times (that differ) on one machine (a laptop, say), then 'all' the machines on the network, especially the file servers, must be configured that way? (I could see this being an issue for timestamps on backups across NFS, and on Dovecot which is very sensitive to time changes.) I've not yet read the tzdata readme, which may discuss some of this, but I will do so, likely after office hours, though. regards, Ron -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: 532c1b04.4020...@tesco.nethttps://lists.debian.org/532c1b04.4020...@tesco.net -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140321135627.19e5f22...@server1.shellworld.net
Re: Time Zone Questions
I chose the posix time for Europe/London and the seconds are in exact step with local time seconds. Martin Ron Leach writes: On 21/03/2014 20:21, John Hasler wrote: Other way around. TAI does *not* include leap-seconds. It is a continuous stream of numbered seconds with no gaps and no insertions. UTC *does* include leap seconds. It is TAI adjusted to stay within one second of Earth rotation time. Leap seconds account for all of the difference between UTC and TAI. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140322004454.826d922...@server1.shellworld.net
Time Zone Questions
What is the difference between the 3 versions of various time zone files? I live in the US-Central time zone and wanted to set a debian system to London time which means replacing /etc/localtime to the file that coresponds to London. That's when I discovered that there are 3 Londons and 3 Chicagos. /usr/share/zoneinfo/right/America/Chicago /usr/share/zoneinfo/America/Chicago /usr/share/zoneinfo/posix/America/Chicago What is the significance of the right and the posix versions of the files? I want to record some radio programs and DST and BST don't start and stop at the same times. This is not an urgent request, but I am curious as to why the 3 versions? The posix version is identical to one of the other 2, at least for Chicago. Thank you. Martin McCormick -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140321023705.b941022...@server1.shellworld.net
