- WLUG - wlug.mailman3.com

Re: [Wlug] last sparc64 gentoo question - I promise
by Jamie Guinan 24 May '07

24 May '07

On Thu, 24 May 2007, ken jones wrote: > I succeeded in doing: > #emerge --sync > #emerge portage > Then I tried > #emerge xorg-x11 > overnight. I woke up to find this had failed on step 101 out of 133. That's actually rather good progress. :) > ------------------- > The first two emerge's completed, emerge xorg-x11 did not. Below is > stdout. Looking at the log it appears that stddef.h is missing... > -----------snip--------- > makedepend: warning: clientattrib.c (reading /usr/include/gconv.h, line > 31): cannot find include file "stddef.h" > ---------snip---------- > Right at the end stdout said, "!!! If you need support, post the > topmost build error, and the call stack if relevant." > I have no idea what this means. Do you? > Any suggestions what I do now? My guess is to just try again. I found this, http://www.nabble.com/emerge-xorg-xll--fails-out-at-media-libs-mesa-6.5-r3-… See the post from Donnie Berkholz near the bottom, where he suggests # eselect opengl set xorg-x11 Then try, # emerge -ptv xorg-x11 [ this will show you what's going to be merged ] # emerge xorg-x11 [ ok, try again...] -Jamie > Ken Jones > ========================================== > -------------------snip of emerge failure step 101 out of > 133---------------- > >>> Emerging (101 of 133) media-libs/mesa-6.5.1-r1 to / > >>> Downloading > 'http://distfiles.gentoo.org/distfiles/MesaLib-6.5.1.tar.bz2' > --00:50:32-- http://distfiles.gentoo.org/distfiles/MesaLib-6.5.1.tar.bz2 > => `/usr/portage/distfiles/MesaLib-6.5.1.tar.bz2' > Resolving distfiles.gentoo.org... 156.56.247.195, 216.165.129.135, > 64.50.236.52, ... > Connecting to distfiles.gentoo.org|156.56.247.195|:80... connected. > HTTP request sent, awaiting response... 200 OK > Length: 3,138,257 (3.0M) [application/x-tar] > > 100%[===============================================================================>] > 3,138,257 523.02K/s ETA 00:00 > > 00:50:38 (503.34 KB/s) - `/usr/portage/distfiles/MesaLib-6.5.1.tar.bz2' > saved [3138257/3138257] > > * checking ebuild checksums ;-) > ... > [ ok ] > * checking auxfile checksums ;-) > ... > [ ok ] > * checking miscfile checksums ;-) > ... > [ ok ] > * checking MesaLib-6.5.1.tar.bz2 ;-) > ... > [ ok ] > * QA Notice: USE Flag 'amd64-fbsd' not in IUSE for media-libs/mesa-6.5.1-r1 > * QA Notice: USE Flag 'kernel_FreeBSD' not in IUSE for > media-libs/mesa-6.5.1-r1 > >>> Unpacking source... > >>> Unpacking MesaLib-6.5.1.tar.bz2 to > /var/tmp/portage/media-libs/mesa-6.5.1-r1/work > * Applying 6.5.1-freebsd-dri.patch > ... > [ ok ] > * Removing DO_DEBUG defs in dri drivers... > * Sparc assembly code is not working; deactivating > >>> Source unpacked. > >>> Compiling source in > /var/tmp/portage/media-libs/mesa-6.5.1-r1/work/Mesa-6.5.1 ... > (cd configs && rm -f current && ln -s linux-dri current) > . > . <<=============many steps in here=========>> > . > clientattrib.c: In function `__glFreeAttributeState': > clientattrib.c:131: error: dereferencing pointer to incomplete type > clientattrib.c:132: error: dereferencing pointer to incomplete type > make[3]: *** [clientattrib.o] Error 1 > make[3]: Leaving directory > `/var/tmp/portage/media-libs/mesa-6.5.1-r1/work/Mesa-6.5.1/src/glx/x11' > make[2]: *** [subdirs] Error 1 > make[2]: Leaving directory > `/var/tmp/portage/media-libs/mesa-6.5.1-r1/work/Mesa-6.5.1/src' > make[1]: *** [default] Error 1 > make[1]: Leaving directory > `/var/tmp/portage/media-libs/mesa-6.5.1-r1/work/Mesa-6.5.1' > make: *** [linux-dri] Error 2 > > !!! ERROR: media-libs/mesa-6.5.1-r1 failed. > Call stack: > ebuild.sh, line 1615: Called dyn_compile > ebuild.sh, line 972: Called qa_call 'src_compile' > ebuild.sh, line 44: Called src_compile > mesa-6.5.1-r1.ebuild, line 215: Called die > > !!! Build failed > !!! If you need support, post the topmost build error, and the call > stack if relevant. > !!! A complete build log is located at > '/var/tmp/portage/media-libs/mesa-6.5.1-r1/temp/build.log'. > > kjones-sun2 bin # > _______________________________________________ > Wlug mailing list > Wlug(a)mail.wlug.org > http://mail.wlug.org/mailman/listinfo/wlug >

1 0

last sparc64 gentoo question - I promise
by ken jones 24 May '07

24 May '07

I succeeded in doing: #emerge --sync #emerge portage Then I tried #emerge xorg-x11 overnight. I woke up to find this had failed on step 101 out of 133. ------------------- The first two emerge's completed, emerge xorg-x11 did not. Below is stdout. Looking at the log it appears that stddef.h is missing... -----------snip--------- makedepend: warning: clientattrib.c (reading /usr/include/gconv.h, line 31): cannot find include file "stddef.h" ---------snip---------- Right at the end stdout said, "!!! If you need support, post the topmost build error, and the call stack if relevant." I have no idea what this means. Do you? Any suggestions what I do now? My guess is to just try again. Ken Jones ========================================== -------------------snip of emerge failure step 101 out of 133---------------- >>> Emerging (101 of 133) media-libs/mesa-6.5.1-r1 to / >>> Downloading 'http://distfiles.gentoo.org/distfiles/MesaLib-6.5.1.tar.bz2' --00:50:32-- http://distfiles.gentoo.org/distfiles/MesaLib-6.5.1.tar.bz2 => `/usr/portage/distfiles/MesaLib-6.5.1.tar.bz2' Resolving distfiles.gentoo.org... 156.56.247.195, 216.165.129.135, 64.50.236.52, ... Connecting to distfiles.gentoo.org|156.56.247.195|:80... connected. HTTP request sent, awaiting response... 200 OK Length: 3,138,257 (3.0M) [application/x-tar] 100%[===============================================================================>] 3,138,257 523.02K/s ETA 00:00 00:50:38 (503.34 KB/s) - `/usr/portage/distfiles/MesaLib-6.5.1.tar.bz2' saved [3138257/3138257] * checking ebuild checksums ;-) ... [ ok ] * checking auxfile checksums ;-) ... [ ok ] * checking miscfile checksums ;-) ... [ ok ] * checking MesaLib-6.5.1.tar.bz2 ;-) ... [ ok ] * QA Notice: USE Flag 'amd64-fbsd' not in IUSE for media-libs/mesa-6.5.1-r1 * QA Notice: USE Flag 'kernel_FreeBSD' not in IUSE for media-libs/mesa-6.5.1-r1 >>> Unpacking source... >>> Unpacking MesaLib-6.5.1.tar.bz2 to /var/tmp/portage/media-libs/mesa-6.5.1-r1/work * Applying 6.5.1-freebsd-dri.patch ... [ ok ] * Removing DO_DEBUG defs in dri drivers... * Sparc assembly code is not working; deactivating >>> Source unpacked. >>> Compiling source in /var/tmp/portage/media-libs/mesa-6.5.1-r1/work/Mesa-6.5.1 ... (cd configs && rm -f current && ln -s linux-dri current) . . <<=============many steps in here=========>> . clientattrib.c: In function `__glFreeAttributeState': clientattrib.c:131: error: dereferencing pointer to incomplete type clientattrib.c:132: error: dereferencing pointer to incomplete type make[3]: *** [clientattrib.o] Error 1 make[3]: Leaving directory `/var/tmp/portage/media-libs/mesa-6.5.1-r1/work/Mesa-6.5.1/src/glx/x11' make[2]: *** [subdirs] Error 1 make[2]: Leaving directory `/var/tmp/portage/media-libs/mesa-6.5.1-r1/work/Mesa-6.5.1/src' make[1]: *** [default] Error 1 make[1]: Leaving directory `/var/tmp/portage/media-libs/mesa-6.5.1-r1/work/Mesa-6.5.1' make: *** [linux-dri] Error 2 !!! ERROR: media-libs/mesa-6.5.1-r1 failed. Call stack: ebuild.sh, line 1615: Called dyn_compile ebuild.sh, line 972: Called qa_call 'src_compile' ebuild.sh, line 44: Called src_compile mesa-6.5.1-r1.ebuild, line 215: Called die !!! Build failed !!! If you need support, post the topmost build error, and the call stack if relevant. !!! A complete build log is located at '/var/tmp/portage/media-libs/mesa-6.5.1-r1/temp/build.log'. kjones-sun2 bin #

2 1

Re: [Wlug] It boots. Now what?
by Jamie Guinan 23 May '07

23 May '07

On Wed, 23 May 2007, ken jones wrote: > I got HAPPYMEAL to work. My sparc64 is on the WEB! > Here is result of ls -lad /etc/make.profile > ----------snip---------- > kjones-sun2 ~ # ls -lad /etc/make.profile > lrwxrwxrwx 1 root root 58 May 17 12:45 /etc/make.profile -> > ../usr/portage/profiles/default-linux/sparc/sparc64/2006.1 > -----------end snip--------- > Here is result of emerge xorg-x11: How do I install from the tar.bz > file? What is gentoo's equivalent of Red Hat's rpm? That's what gentoo's portage system (via the "emerge" command) does for you - downloads the tarballs, builds, installs, and keeps a record of what's installed under /var/db/pkg/<category>/<package>/ . Well, when it works. :) Look at /usr/portage/media-libs/libpng/*.ebuild, its fairly easy to get the gist of it if you know shell syntax. Did you do "emerge --sync" like some others suggested? I see a few libpng versions available, up to 1.2.16, $ ls /usr/portage/media-libs/libpng/*.ebuild /usr/portage/media-libs/libpng/libpng-1.2.14-r1.ebuild /usr/portage/media-libs/libpng/libpng-1.2.14.ebuild /usr/portage/media-libs/libpng/libpng-1.2.15.ebuild /usr/portage/media-libs/libpng/libpng-1.2.16.ebuild > --------------snip------------- > Lastly, how do you know that HAPPYMEAL is in the sun-hme module? Where > is the information indicating the functionality contained in each module? In the "make menuconfig" [Help] items. Also in menuconfig, you can search all the help text by typing "/" and entering some text. Generally, once you have your system configured to your satisfaction, you keep a copy of the kernel .config, and use that the next time you upgrade the kernel by copying it to the new kernel tree and doing "make oldconfig". -Jamie

1 0

It boots. Now what?
by ken jones 23 May '07

23 May '07

I would like to get X11 services. I did an "emerge xterm". Here's what happened. ------------start snip------------- (chroot) livecd proc # emerge xterm Calculating dependencies | !!! All ebuilds that could satisfy ">=x11-proto/xproto-7.0.6" have been masked. !!! One of the following masked packages is required to complete your request: - x11-proto/xproto-7.0.7 (masked by: ~sparc keyword) - x11-proto/xproto-7.0.10 (masked by: ~sparc keyword) - x11-proto/xproto-7.0.9 (masked by: ~sparc keyword) For more information, see MASKED PACKAGES section in the emerge man page or refer to the Gentoo Handbook. (dependency required by "x11-libs/libX11-1.1.1-r1" [ebuild]) --------------end snip---------- What do I make of this? Another problem. My livecd prompt and my chroot prompt both see my ethernet card, eth0. However my new boot does not see the card. Ideas? Ken Jones

3 3

I got my sparc64 gentoo on the network
by ken jones 23 May '07

23 May '07

Thank you for suggesting that I compare two dmesg boot message outputs. "grep eth0" on the dmesg outputthat worked referenced HAPPYMEAL. The same 'grep' on the boot log that did not work had no HAPPYMEAL string. I had a 'CONFIG_HAPPYMEAL=m' line in .config. I changed it to 'CONFIG_HAPPYMEAL=y' and recompiled using genkernel. NOW MY NEWLY BOOTED sparc64 IS NETWORKED. The sky's the limit. :-) What I do not understand is how do I know which Module contains which functionality. If I had stuck with the '=m' config I should be able to add a module at boot time. How do I know which module to add? Ken Jones

3 2

gentoo on sparc64 portage question..
by ken jones 23 May '07

23 May '07

Please someone explain to me what is happening here. ---------------snip--------------------- (chroot) livecd dhcpcd # emerge dhcpcd Calculating dependencies... done! >>> Emerging (1 of 1) net-misc/dhcpcd-3.0.16-r1 to / >>> checking ebuild checksums ;-) >>> checking auxfile checksums ;-) >>> checking miscfile checksums ;-) >>> checking dhcpcd-3.0.16.tar.bz2 !!! Missing digest for 'dhcpcd-3.0.16.tar.bz2' (chroot) livecd dhcpcd # ---------------end snip---------------- Is this a successful emerge? If not, what do I do about the missing digest? Is there an emerge parameter that eliminates the sum check? I looked in 'man emerge' to no avail. emerge is supposed to install things. Instead of installations I get whatever this is. I can not "find" dhcpcd (a binary or a script) anywhere inside my machine. Is another step needed? Ken Jones

3 2

Re: [Wlug] PXE 101 (summer school)
by Adam Keck 22 May '07

22 May '07

<oops... replied off list, reposting> http://www.kernel.org/pub/linux/utils/boot/syslinux/ has the PXE server source that most distros use and http://syslinux.zytor.com/pxe.php talks about how to setup PXE, tftpd, and dhcpd (ISC). -Adam On 5/22/07, Doug Mildram <doug.mildram(a)mindspeed.com> wrote: > > doug> I too use PXE at work, with some success. > I could stand to learn more, and I sure as hell am still > looking for the "pxe daemon source code"...buried by intel?who? > while still running a redhat binary pxe daemon here. > > If every box had a bootable CD, it'd be easier to > just do that ...whether the CD is the stock CD, or one that > you customize a bit, which I've done; which is separate from > customizing the kickstart config file... > but realize as an opening comment, I may recommend that you > ignore PXE until you need it; ignore the boot-from-NIC potential. > > You're VERY welcome to torture me with questions, or visit Westboro > (work/days..it's quiet enough here) for some hands on confusion. > > It may be easier to LEARN PXE after learning "kickstart" without PXE. > e.g. booting from CD with > "linux ks=nfs:nfsserverip:/path/to/yourconfigfile.cfg" > and the .cfg file has MORE info like > ....some intro omitted..... > network ... (static,dhcp...it can install static > while USING a temp-dhcp installaddr) > part .... (disk layout) > authconfig ... ( --enablenis is my option) > nfs --server servername-or-ip --dir /path/releasedir > skipx (no X, dumb terminal consoles best for servers, right?) > %packages... > ...... > % post > exec > /KICKSTART.log 2>&1 > echo "hello..." > mkdir /customdir > mount nfsserver:/exporteddir /customdir > ( grab things, even edit files in / ....) > chkconfig --level 345 ntpd on > chkconfig --level 345 nfs on > ..............and so on. Tends to be a few hundred lines long. > > PXE does all this, but without the CD...and with more flaky results > depending on NIC cards, mostly, i think...I find it > just complicated enough to confuse me, > and just odd enough to NOT work in some cases. > > Challenges: 1) can't find the source to the pxe daemon, > which seems to have fallen OFF of the distros. > I feel "stuck" running a redhat9 binary (which works usually-for-me > on a redhat3-or-4 based system, just copying /usr/sbin/pxe > and firing it up manually (it often crashes after getting its job done) > > 2) complex dependencies. It needs > 2a) DHCP server, can be your existing one > 2b) TFTP server (this part is NOT complex, after you > learn some of the bootfile/dir layout) > as well as the (nfs,ftp,youdecide) server to grab packages from. > > What's confusing is how the PXE daemon and DHCP daemon play together. > When my boot fails, the error whines about DHCP even when PXE is the > problem. > >

1 0

Re: [Wlug] It boots. Now what?
by sageman＠WPI.EDU 22 May '07

22 May '07

> Message: 1 > Date: Tue, 22 May 2007 09:16:18 -0400 > From: ken jones <kjones(a)ziplink.net> > Subject: [Wlug] It boots. Now what? > To: WLUG <wlug(a)mail.wlug.org> > Message-ID: <4652ED22.5090802(a)ziplink.net> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > I would like to get X11 services. I did an "emerge xterm". Here's what > happened. > ------------start snip------------- > (chroot) livecd proc # emerge xterm > Calculating dependencies | > !!! All ebuilds that could satisfy ">=x11-proto/xproto-7.0.6" have been > masked. > !!! One of the following masked packages is required to complete your > request: > - x11-proto/xproto-7.0.7 (masked by: ~sparc keyword) > - x11-proto/xproto-7.0.10 (masked by: ~sparc keyword) > - x11-proto/xproto-7.0.9 (masked by: ~sparc keyword) > > For more information, see MASKED PACKAGES section in the emerge man page > or > refer to the Gentoo Handbook. > (dependency required by "x11-libs/libX11-1.1.1-r1" [ebuild]) > --------------end snip---------- > > What do I make of this? > > Another problem. My livecd prompt and my chroot prompt both see my > ethernet card, eth0. However my new boot does not see the card. Ideas? > > Ken Jones Ken, Pulling in xterm, it needs libX11, which, in turn, is pulling in xproto. For the sparc, a masked version of xproto seems to be required. This masking with the "~" is just saying it's in the "testing" set of packages, but it's not hardmasked (it would mention it if it was), meaning it *should* be stable. Regardless, you need the "unstable" version in order to get xterm. There's a number of ways to deal with this, but the suggested way is to add the necessary files, one by one, to /etc/portage/package.unmask. You can do this by adding a line with "x11-proto/xproto ~sparc", or, even better, using the flagedit utility (which you must first "emerge flagedit" to get), and just do: "flagedit x11-proto/xproto +~sparc". [http://damz.net/flagedit/] More info on masking: http://gentoo-wiki.com/Masked

1 0

Re: [Wlug] [Fwd: Re: Multiple RS232 Com Ports...]
by Stephen Daukas 22 May '07

22 May '07

Hi all! Lots of good info... I hadn't stopped to think about a terminal server! OK, lets see about answering some of the questions... - - - - - - Yes, cabling will be fun. The sondes are used for sampling water... The sondes we use are cylinders approximately 2.5 feet in length and about 3 inches in diameter (one, called "big Bertha", is about six inches in diameter) . One end has a data connection that uses a round, screw-on, water tight serial connector, and the business end has several sensors sticking out (optical, chemical, conductivity, a little propeller for mixing water to be sampled, etc., etc.), all within a screw-on cage that protects the sensors from stuff like rocks passing by in the current (fast currents equate to rather large rocks). The idea is to have a tank of water (fish tank, cooler, whatever) fitted with Plexiglas that has multiple holes through which the several sondes are positioned upright, data end pointing up, with a bunch of cables hanging in-place above each sonde - think of the device that sets the bowling pins in place at the end of the alley... Anyway, you place a bunch of sondes into the tank (that has circulating water being maintained at a precise temp, % dissolved Oxygen, and so on), plug each sonde into the cable available directly above it, and then go to the *insert Linux solution here* and initiate communication, data upload, and calibration routines. (By the way, the tank is just about done - its is being debugged as we speak.) After upload and calibration (using the water in the tank), configure each sonde for the next deployment (how often to sample, etc.), turn them off, unscrew the cables and screw on a waterproof cover, remove them from the tank, and place the entire sonde into another protective chamber (6" PVC pipe with holes drilled in a specific pattern and capped on both ends) for another few weeks in some river/lake/coastal location. - - - - - - Yes, I may be confusing myself regarding "I/O device driver" versus "driver" for the sonde... However, I'm pretty sure that all I need is a vanilla "RS232 device driver" that will "just work" for any number of "com" ports by allowing some vanilla terminal emulation package (preferably one that has scripting capabilities) to talk to multiple RS232 ports. I could be wrong about this, in which case I may need to write some code, or perhaps the manufacturer might be thinking of supporting Linux. I don't do the day-to-day instrumentation work, but I am supposed to worry about it none the less. In terms of the "sondes going to sleep", this is based on concerns expressed by the lab along the lines of "if you hook-up the sonde and let it sit too long, you have to start over because it will turn itself off." Also, I use the term "vanilla" a lot because the staff aren't computer types - this has to "just work" like any other tool in the lab. Dear lord - we are talking about biology types after all! ;-) (Unlike engineering/geoscience types!!) - - - - - - So, the other question was about the necessity of having certain signals present. Again, I'm told that the sonde needs all the signals present in a DB-9. I'm guessing this may have more to do with not really knowing for sure and erring on the side of caution, but since there is hardware available that will do this and I can't spend all of my time chasing this down, I'll assume the information I'm getting on this topic is correct. (Biologists not withstanding.) I think this is where the multiport boards or a terminal server would come in... One of the replies mentioned writing a script, and that is what I was thinking about... Specifically, I was wondering if it could be as simple as a script sending the proper bit patterns to get the sonde to do the right things, pipe the serial data through the script and append the data to a file, and s on... (Was that you, John?) - - - - - - - Anyway, I'll have to take a look at the product references mentioned and contemplate what I might need to do in the world of scripting... I haven't heard back from the manufacturer yet, but I'm hoping this will turn out to be a simple project when all is said and done!! Also, I haven't used a terminal server before, so that should be interesting as well. Thanks! If you have any more thoughts, please let me know!! Steve On 5/21/07, Stephen C. Daukas <scd(a)daukas.com> wrote: > > > > -------- Original Message -------- Subject: Re: [Wlug] Multiple RS232 > Com Ports... Date: Mon, 21 May 2007 15:44:09 -0400 From: John Stoffel > <john(a)stoffel.org> <john(a)stoffel.org> Reply-To: Worcester Linux Users > Group <wlug(a)mail.wlug.org> <wlug(a)mail.wlug.org> To: Worcester Linux Users > Group <wlug(a)mail.wlug.org> <wlug(a)mail.wlug.org> References: > <4650752A.1020203(a)daukas.com> <4650752A.1020203(a)daukas.com> > <18001.54171.525952.976484(a)stoffel.org><18001.54171.525952.976484(a)stoffel.org> > <20070521181244.GH4900(a)angus.ind.WPI.EDU><20070521181244.GH4900(a)angus.ind.WPI.EDU> > > Chuck> On Mon, May 21, 2007 at 01:15:07PM -0400, John Stoffel wrote: > >> darn good things about the Cyclades stuff as well. Stay away from the > >> Legacy Avocent (who bought Cyclades...) serial port stuff, it's crap. > > Chuck> Hmm I have to disagree here. We have the Avocent/Cyclades ACS line > Chuck> and are very happy with them. They run Linux inside and support many > Chuck> connection protocols, like Telnet, SSH, Reverse Telnet, Bidirectional, > Chuck> etc. Full RTS/CTS control lines including DTR are available. They > Chuck> support console logging to RAM, PCMCIA flash card, or remote syslog. > > Chuck> Perhaps you are referring to their older Cyclades TS line. > > I'm talking about the old Avocent designed CPS line of crap. You're > talking about the Cyclades designed stuff, including the TS and the > ACS. > > I've got an ancient Cyclom-Y ISA 8-port serial card in my main home > computer and it's been working great for for years. Wish I could > afford the newer stuff. > > John > > _______________________________________________ > Wlug mailing list > Wlug(a)mail.wlug.org > http://mail.wlug.org/mailman/listinfo/wlug > >

2 1

Re: [Wlug] Setting up PXE/TFTBoot
by Mike Leo 22 May '07

22 May '07

no, but i bet these guys have some doc's for you http://www.ltsp.org/ ----- Original Message ---- From: Tal Cohen <wlug(a)cohen123.com> To: wlug(a)mail.wlug.org Sent: Tuesday, May 22, 2007 9:34:32 AM Subject: [Wlug] Setting up PXE/TFTBoot  I know that this is not new technology….but I am having quite a time figuring this out. I was hoping that you guys could give me a hand. In my office, I want to setup a system that will provide a boot image over the network so that I can remotely setup and configure new servers. I am running Crux2.3. Does anyone have experience setting up remote booting systems? Thanks, Tal _______________________________________________ Wlug mailing list Wlug(a)mail.wlug.org http://mail.wlug.org/mailman/listinfo/wlug

1 0

Setting up PXE/TFTBoot
by Tal Cohen 22 May '07

22 May '07

I know that this is not new technology..but I am having quite a time figuring this out. I was hoping that you guys could give me a hand. In my office, I want to setup a system that will provide a boot image over the network so that I can remotely setup and configure new servers. I am running Crux2.3. Does anyone have experience setting up remote booting systems? Thanks, Tal

2 1

Multiple RS232 Com Ports...
by Stephen C. Daukas 21 May '07

21 May '07

Hi all! I'm wondering if anyone knows how many RS232 devices a "vanilla" Linux kernel can manage simultaneously. I'd like to be able to have up to 32 going at once, with data taken from each device written to its own file via the standard I/O subsystem. As far as I know, wouldn't Linux simply fork the appropriate number of driver threads to read as many RS232 ports as are attached with each device's stream written to a "/dev" file? Why do I want to do this? Glad you asked... I'm investigating an issue at work that involves our instrumentation lab having to upload/download/calibrate multiple sondes and have them ready for field deployment the next day. Currently, we can do three at a time and this takes about 8-9 hours providing everything goes well. The manufacturer's software application claims to supports the simultaneous upload/config of up to 32 sondes. The application will scan "installed" COM ports looking for recognizable sondes, and present those found to the user. Yes - this is a windows solution, so I first began looking for products that would play with Windows... I began looking for hardware solutions (PCI boards, break-out boxes, etc.) on the net and I found a few that will provide 16 DB-9 RS232 ports with full pin-out support as well as so-called "true" interrupt and FIFO support on the host (via their own custom Windows driver, no doubt). And then I started wondering about how to get all of this to work in a Windows environment where "IT" is the only group allowed to install software and otherwise configure a PC for most users whose biggest concern is email... All this got me thinking about avoiding the whole Windows issue and instead thinking about using Linux to take care of the RS232 part, with the newly created files moved to a Windows box (sneakernet to start with), with the normal post-processing of the data taking place as usual in the Windows environment the staff is used to... I'm hoping an "out of the box" distro will be able to handle say 16 RS232 ports and that Linux-based terminal emulation software is available to simultaneously handle those ports as well so I can capture the data using the appropriate protocols... Anyone know of a similar application using RS232 and Linux, or have some thoughts on this? Thanks! Steve

6 8

My Ultra sparc64 booted!!
by ken jones 21 May '07

21 May '07

Thank you everybody for your help. My sparc64 finally booted. I used genkernel with the livecd cdrom's kernel 2617 /proc/config.gz. The last piece of the puzzle was getting fstab correct. Here is proof. -----------snip of uname -a ----------------------- kjones-sun2 / # uname -a Linux kjones-sun2 2.6.20-gentoo-r5 #3 SMP Thu May 17 13:58:22 Local time zone must be set--see zic sparc64 sun4u GNU/Linux kjones-sun2 / # -------------end snip-------------- I am not sure whether I learned anything specific to gentoo or to sparc64. I did learn that creating good open source documentation is difficult and that finding one's way through unfamiliar documentation is daunting. Thanks, again. Ken Jones

1 0

Fwd: Urgent Need for Software QA Instructor in Boston, MA
by albutler33＠netscape.net 20 May '07

20 May '07

Hello all, Here's an opportunity I thought I'd pass along. Al Butler member of WLUG (haven't attended for quite a while) -----Original Message----- From: april01(a)o-s-tech.com To: albutler33(a)netscape.net Sent: Thu, 17 May 2007 8:59 PM Subject: Urgent Need for Software QA Instructor in Boston, MA Dear Aljerin, Open Systems Technology Services has an urgent need for an experienced Software Quality Assurance Instructor. Topics Software development life cycle-the product view Software configuration management (SCM) Software quality assurance Testing, validating and evaluating Special topics Planning, organizing, implementing and managing for SQA ------------------- Location: Boston, MA Date: June 11-13 Rate: $300 day/ 1099 Please let me know ASAP if you are available or know of someone who have experience teaching Linux. Classes will begin at 8am each day. Introduction to Linux, 3 days March 19th - 21st Linux Admin I, 5 days March 26th - 30th Linux Admin II, 5 days April 9th - 13t Rate: $500.00/day Job Description: . If you are available, email your resume (in word format) and availability with a contact number by 5:00pm, Monday, May 21, 2007. Thank you, April Washington Open Systems Technology Services, Inc. Email: April01(a)o-s-tech.com ________________________________________________________________________ Check Out the new free AIM(R) Mail -- 2 GB of storage and industry-leading spam and email virus protection.

1 0

ALSA hanging (Gentoo AMD64)
by Eric Stein 18 May '07

18 May '07

Hi all, I have a very strange problem. Last night I updated all my packages and rebooted. On boot, alsasound started (boot runlevel), but while trying to restore mixer levels, it hung. I had to boot to a livecd and remove alsasound from boot to even start my system. I tried removing the old state file (in case the format changed and it was breaking because of that), but it wouldn't start without it. glamdring ~ # /etc/init.d/alsasound start * One of the files in /etc/{conf.d,init.d} or /etc/rc.conf * has a modification time in the future! * Loading ALSA modules ... [ ok ] * Restoring Mixer Levels ... * No mixer config in /var/lib/alsa/asound.state, you have to unmute your card! [ ok ] glamdring ~ # cd /var/lib/alsa/ glamdring alsa # l total 9 drwxr-xr-x 3 root root 152 May 17 16:26 ./ drwxr-xr-x 23 root root 568 May 17 03:17 ../ -rw-r--r-- 1 root root 0 May 17 16:26 .keep_media-sound_alsa-utils-0.9 -rw-r--r-- 1 root root 8073 May 17 12:50 asound.state.old drwxr-xr-x 2 root root 144 May 17 12:50 oss/ glamdring alsa # mv asound.state.old asound.state glamdring alsa # cd glamdring ~ # /etc/init.d/alsasound zap * Manually resetting alsasound to stopped state. glamdring ~ # /etc/init.d/alsasound start * Loading ALSA modules ... [ ok ] * Restoring Mixer Levels ... Which is where it hangs. Other alsa programs like alsaconf and alsamixer hang on startup, along with programs that attempt to use ALSA to output sound. I've tried reinstalling ALSA, I've tried revdep-rebuild, I've done modules-rebuild, and I'm at my wit's end. None of it seems to make sense, and Gnome isn't very happy about not having a sound output, so half of my GUI is missing. If anyone has some idea why this could be happening to me, please let me know. Eric

1 0

OT: HP9k E-class
by Desdinova 17 May '07

17 May '07

A bit OT, if you can reply off-list, that's fine. I just figure there's probably guys who know about crappy old minis here. I've got an E25 and I'm planning to replace the internal scsi drive, it shouldn't be any big deal, but this *is* HP. Are there any gotchas on drive geometry or size that I should watch for? Take care, des

1 0

Re: [Wlug] Follow-Up to Key Signing Party
by Jamie Guinan 17 May '07

17 May '07

On Thu, 17 May 2007, Eric Martin wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Thanks to everyone who showed up to tonight's key signing party. Now > that we verified everybody had the correct key info and identification, > we actually have to sign each other's keys. We are using the keyserver > pgp.mit.edu. If that isn't your default, you can set it either in > ~/.gnupg/gpg.conf or at the command line for each command by using the > argument > - --keyserver pgp.mit.edu Is the "hkp://" prefix needed in gpg.conf? keyserver hkp://pgp.mit.edu > Below is an excerpt from the gpg Key Signing Party HowTo that explains > things in detail.(http://www.rubin.ch/pgp/kspa/gpg-party.en.html#ss3.8): > > 3.8 How to sign others' keys > > Step 1: Get a copy of the key > > Normally, you'll be working from a keyserver. However if you are signing > the key that is not available on a keyserver, you can use simply import > the key with gpg --import. If you are working with a keyserver, the > following command will download the key from the keyserver into your > public keyring. > > bash$ gpg --keyserver <keyserver> --recv-keys <Key_ID> > > If you get a read error, it means the keyserver is overloaded. Please, > try again in a few seconds. That worked Ok. > > Step 2: Fingerprint and Verify the key > > bash$ gpg --fingerprint <Key_ID> That too. > GPG will print out the fingerprint of the Key with <Key_ID > (the key > you just downloaded). Check the fingerprint against the checklist that > you where given at the party. Note: Don't check the fingerprint on your > checklist against the fingerprint on the web page as the server may not > send you the same key it displays on the web page. > > Step 3: Sign the key > > bash$ gpg --sign-key <Key_ID> Ok. > > If you have multiple private keys, you can specify which of your private > keys to sign the other persons public key with like this: > > bash$ gpg --default-key <Key_to_use> --sign-key <Key_ID> > > Step 4: Return or Upload the signed key > > If you are working with an entity which does not want their key on a > public keyserver, you should at this point you should return their > signed key back to them by their method of choice - normally encrypted > email. You should not send a public key to a keyserver with out the > permission of the key's owner. Publicizing a public key slightly reduces > the security of a key pair, therefor it is considered rude to make a key > more public than its owner desires. > > Most likely you are working with a keyserver. If that is the case, you > can send the signed key back to the keyserver like this: > > bash$ gpg --keyserver <keyserver> --send-key <Key_ID> > > You should see a success message like this: > > gpg: success sending to `<keyserver>' (status=200) > Congratulations, the signature of the other entity's key is now complete > and your signature has been incorporated into their public key. A trust > path has been established. This looked different for me, $ gpg --send-key A9413B9F gpg: sending key A9413B9F to hkp server pgp.mit.edu $ No affirmative response, it just returned to the prompt. But the exit code was 0. $ echo $? 0 -Jamie [ not yet using gpg for mail ]

1 0

Follow-Up to Key Signing Party
by Eric Martin 17 May '07

17 May '07

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Thanks to everyone who showed up to tonight's key signing party. Now that we verified everybody had the correct key info and identification, we actually have to sign each other's keys. We are using the keyserver pgp.mit.edu. If that isn't your default, you can set it either in ~/.gnupg/gpg.conf or at the command line for each command by using the argument - --keyserver pgp.mit.edu Below is an excerpt from the gpg Key Signing Party HowTo that explains things in detail.(http://www.rubin.ch/pgp/kspa/gpg-party.en.html#ss3.8): 3.8 How to sign others' keys Step 1: Get a copy of the key Normally, you'll be working from a keyserver. However if you are signing the key that is not available on a keyserver, you can use simply import the key with gpg --import. If you are working with a keyserver, the following command will download the key from the keyserver into your public keyring. bash$ gpg --keyserver <keyserver> --recv-keys <Key_ID> If you get a read error, it means the keyserver is overloaded. Please, try again in a few seconds. Step 2: Fingerprint and Verify the key bash$ gpg --fingerprint <Key_ID> GPG will print out the fingerprint of the Key with <Key_ID > (the key you just downloaded). Check the fingerprint against the checklist that you where given at the party. Note: Don't check the fingerprint on your checklist against the fingerprint on the web page as the server may not send you the same key it displays on the web page. Step 3: Sign the key bash$ gpg --sign-key <Key_ID> If you have multiple private keys, you can specify which of your private keys to sign the other persons public key with like this: bash$ gpg --default-key <Key_to_use> --sign-key <Key_ID> Step 4: Return or Upload the signed key If you are working with an entity which does not want their key on a public keyserver, you should at this point you should return their signed key back to them by their method of choice - normally encrypted email. You should not send a public key to a keyserver with out the permission of the key's owner. Publicizing a public key slightly reduces the security of a key pair, therefor it is considered rude to make a key more public than its owner desires. Most likely you are working with a keyserver. If that is the case, you can send the signed key back to the keyserver like this: bash$ gpg --keyserver <keyserver> --send-key <Key_ID> You should see a success message like this: gpg: success sending to `<keyserver>' (status=200) Congratulations, the signature of the other entity's key is now complete and your signature has been incorporated into their public key. A trust path has been established. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGS+VsaiVxdKlBO58RAs3xAJ0W7SyjfbQtwHwDaHeWG5EgeWWXzgCfQ8pT W40gnhPVnkQ3u7E1jbXALvU= =GM/c -----END PGP SIGNATURE-----

1 0

PGP Key Signing Party
by Eric Martin 17 May '07

17 May '07

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Just a reminder that we are having a PGP Key Signing Party tomorrow for WLUG. If you're name isn't listed below, I do not have your info (and you should get it to me if you want to be on the list). I know Andy said noon tomorrow but I'll take stuff until 4pm. Don't forget your two forms of ID and I'll see you tomorrow. Eric Martin Gregory M. Avedissian Walt Sawyer Alexander Pennace Doug Mildram Keith Wright Alex Weiss Andrew M. Stewart James Philip Gray - -Eric -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGSj+vaiVxdKlBO58RAmxgAJ0QaQlNCSrw58e80VeQjjGL/GKsOwCdErcn vMjPw9eiQD4zXEzjNWPBKa4= =KiW+ -----END PGP SIGNATURE-----

4 3

Last-Minute Emergency, Unable to Attend
by Terry S 17 May '07

17 May '07

Hello, I had a phone call at 6:07 that required me to go out for a while, and did not have time to email. (I'm the one with the camera.) Hopefully, someone else just "happened" to have one, but I can take pictures at our next meeting if needed. How did the key-signing go? Terry

1 0

large file performance on linux
by brad noyes 16 May '07

16 May '07

Hello All, I am seeing some really slow performance regarding large files on linux. I write a lot of data points from a light sensor. The stream is about 53 Mb/s and i need to keep this rate for 7 minutes, that's a total of about 22Gb. I can sustain 53Mb/s pretty well until the file grows to over 1Gb or so, then things hit the wall and the writes to the filesystem can't keep up. The writes go from 20ms in duration to 500ms. I assume the filesystem/operating system is caching writes. Do you have any suggestions on how to speed up performance on these writes, filesystem options, kernel options, other strategies, etc? Things I have tried: - I have tried this on a ext3 file system as well as an xfs filesystem with the same result. - I have also tried spooling over several files (a la multiple volumes) but i see no difference in performance. In fact, i think this actually hinders performance a bit. - I keep my own giant memory buffer where all the data is stored and then it is written to disk in a background thread. This helps, but i run out of space in the buffer before i finish taking data. Thanks, -- Brad

6 13

Meeting tomight
by freak4uxxx＠gmail.com 16 May '07

16 May '07

I might be slightly delayed tonight, small family thing that needs to be taken care of. I shouldn't be late but I just wanted to let people know. -Eric

2 1

Reminder: Upcoming PGP Key Signing Party
by Andy Stewart 16 May '07

16 May '07

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 HI Everybody, At our next WLUG meeting on Wednesday, May 16th, we will have a PGP Key Signing Party hosted by Chuck Anderson and Eric Martin. Eric has previously sent out instructions, so please be sure to follow them, and direct any questions to Eric or Chuck. I have posted those instructions on the WLUG home page (see description for May WLUG Meeting). I suggest a cutoff of noon on Wednesday, May 16th, for sending your information to Eric. This will give Eric time to collect the necessary information prior to the meeting. Dust off those passports - you'll need them! Later, Andy - -- Andy Stewart, Founder Worcester Linux Users' Group (http://www.wlug.org) Chelmsford Linux Meetup Group (http://linux.meetup.com/393) Amateur Radio: KB1OIQ -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org iD8DBQFGQmjGHl0iXDssISsRAr1rAJ91nX/DNXHjRie+dDYyVjrq8ePq2wCeN8DC Q+SoklLvcW6MfOaCGuvwKwU= =kUPV -----END PGP SIGNATURE-----

4 3

Re: [Wlug] large file performance on linux
by Jamie Guinan 16 May '07

16 May '07

On Wed, 16 May 2007, brad noyes wrote: > Hello All, > > I am seeing some really slow performance regarding large files on linux. I > write a lot of data points from a light sensor. The stream is about 53 Mb/s and > i need to keep this rate for 7 minutes, that's a total of about 22Gb. I > can sustain 53Mb/s pretty well until the file grows to over 1Gb or so, then > things hit the wall and the writes to the filesystem can't keep up. The writes > go from 20ms in duration to 500ms. I assume the filesystem/operating system > is caching writes. Do you have any suggestions on how to speed up performance > on these writes, filesystem options, kernel options, other strategies, etc? > > Things I have tried: > - I have tried this on a ext3 file system as well as an xfs filesystem > with the same result. > > - I have also tried spooling over several files (a la multiple volumes) > but i see no difference in performance. In fact, i think this actually > hinders performance a bit. > > - I keep my own giant memory buffer where all the data is stored and > then it is written to disk in a background thread. This helps, but > i run out of space in the buffer before i finish taking data. Hi, Do you mean MB/s as in MegaBytes/second? Lower-case "b" implies "bits" to me, but I suspect you meant bytes. :) Anyway, if you know your total data set is going to exceed your system memory, which will largely get used as page cache, you might as well open the output with O_SYNC and write straight out. Then the big buffer in the writer thread can go away, and you can just queue chunks between the input and writer. A good experiment would be to run the writer thread with dummy input data and see what kind of throughput you get. You could try ext2 (no journalling), or tweaking some of the journalling options in ext3 (data=journal/ordered/writeback, see "man mount"). If you have enough CPU bandwidth, and your input stream has enough redundancy, you could gzip it before writing, which might reduce your output bandwidth requirements. Hope this helps, keep us posted. -Jamie > > Thanks, > -- Brad > > _______________________________________________ > Wlug mailing list > Wlug(a)mail.wlug.org > http://mail.wlug.org/mailman/listinfo/wlug >

2 1

SUN Ultra1 booting (still, sorry)
by ken jones 08 May '07

08 May '07

I used genkernel to compile my 2.6.20-gentoo-r5 kernel. After much thrashing I got it to boot. All looked good until the very end when it claimed that it "Failed to set the user font" announced the name of my machine and gave me the login prompt. It would not respond to any keyboard input. After a minute or two it printed a line about "disabled for 5 minutes" and went permanently dead. See below. Any ideas? ---------------snip from almost successful boot--------------- * Starting lo [ !! * Bringing up lo * 127.0.0.1/8 * Adding routes [ ok * 127.0.0.0/8 ... * Initializing random number generator ... [ ok NIT: Entering runlevel: 3 [ ok * Setting user font ... putfont: PIO_FONT: Function not implemented putfont: PIO_FONT: Function not implemented putfont: PIO_FONT: Function not implemented putfont: PIO_FONT: Function not implemented putfont: PIO_FONT: Function not implemented putfont: PIO_FONT: Function not implemented putfont: PIO_FONT: Function not implemented putfont: PIO_FONT: Function not implemented putfont: PIO_FONT: Function not implemented putfont: PIO_FONT: Function not implemented putfont: PIO_FONT: Function not implemented * Failed to set user font * Mounting network filesystems ... [ !! * Starting local ... [ ok [ ok This is kjones-sun2.unknown_domain (Linux sparc64 2.6.20-gentoo-r5) 21:42:42 kjones-sun2 login: INIT: Id "s0" respawning too fast: disabled for 5 minutes

1 0