Re: SL44 and sun fire x4600
No sorry. On Thu, 15 Mar 2007, Martin Flemming wrote: Hi, Stephen ! Yep, this was also my thought, but this kernel "kernel-largesmp-2.6.9-42.0.3.EL.x86_64" crashes as i remarked ... Any other ideas ? cheers, Martin On Thu, 15 Mar 2007, Stephen J. Gowdy wrote: It looks like you need largesmp (assuming you have the 8 dual-core CPU version, most option look to only include 4 dual-core CPUs); "Please note that limits for v4 are for Update 3 or later. Update 3 was released in March 2006. CPU counts over 8 (AMD64/EM64T) or 64 (other architectures) require use of the largesmp kernel. Certified limits reflect the current state of system testing by and its partners, and the limit of support provided by a Linux subscription." On Thu, 15 Mar 2007, Martin Flemming wrote: Hi, Troy et all ! I've recognized today, that on the hardware-webside https://www.scientificlinux.org/documentation/hardware/ you've published the sucessfull installation of a "Sun Fire x4600"-Machine ... We've got the same machine in our lab, but unfortunatley we see only 8 cpu's not 16 ... So my question is, which kernel do you have installed ? I've installed following one: kernel-smp-2.6.9-42.0.3.EL.x86_64 which displays only 8 CPU's ... At first, i've got the largesmp-kernel kernel-largesmp-2.6.9-42.0.3.EL.x86_64 but this one generate a kernelpanic Cheers, Martin __ Martin Flemming DESY / IT office : Building 2b / 008a Notkestr. 85 phone : 040 - 8998 - 4667 22603 Hamburg mail : [EMAIL PROTECTED] __ -- /--------+-\ |Stephen J. Gowdy, SLAC | CERN Office: 32-2-A22| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| |http://calendar.yahoo.com/gowdy | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/ Gruss Martin Flemming __ Martin Flemming DESY / IT office : Building 2b / 008a Notkestr. 85 phone : 040 - 8998 - 4667 22603 Hamburg mail : [EMAIL PROTECTED] __ -- /--------+-\ |Stephen J. Gowdy, SLAC | CERN Office: 32-2-A22| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| |http://calendar.yahoo.com/gowdy | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/
RHEL5 question...
Hi All, Sorry for the off topic question but thought I might get an answer quicker here. I'm trying to install RHEL5 on a machine (currently running RHEL3) it has trouble reading the partition table somehow. I can read it fine with fdisk after pushing ALT-F2 to get the shell up. Is there a limit in the number of partitions anaconda understands? Here is the partition table; Disk /dev/hde: 40.0 GB, 40020664320 bytes 255 heads, 63 sectors/track, 4865 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Device BootStart EndBlocks Id System /dev/hde1 * 133265041 83 Linux /dev/hde234 4865 388130405 Extended /dev/hde5 1107 4865 30194136 83 Linux /dev/hde6 845 1106 2104515 83 Linux /dev/hde734 582 4409779+ 83 Linux /dev/hde8 583 713 1052226 83 Linux /dev/hde9 714 844 1052226 82 Linux swap Originally it was out of order but I fixed that and it still has the problem. The message starts "partition table on device hda was unreadable" (this device is hda during the installation), it wants you to create a new partition table. I'd rather not as I want to keep some partitions after the reinstall. Any ideas? regards, Stephen. -- /+---------\ |Stephen J. Gowdy, SLAC | CERN Office: 32-2-A22| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| |http://calendar.yahoo.com/gowdy | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/
Re: RHEL5 question...
[my first reply to you bounced from the list as I sent it "from" CERN instead of SLAC] I just went to try turning off the RAID controller in the BIOS to see if that would help but after going through all the options I remembered you need to set a jumper to do that. I might just let it wipe out the partition table, I'll copy the data I wanted to keep to another disk (the one on the RAID controller) and then copy it back after the install. On Tue, 20 Mar 2007, Oleg Sadov wrote: Looks like improperly configuration of multipath software-RAID. For example, changed or not properly configured drivers, incorrect device path or something else. Simplest way for resolving of this problem -- disconnecting all suspicious devices, installing system to something common disk drive (IDE, SCSI or SATA) and then carefully set up your RAID-subsystem for current version of system. Of course, output of sysreport utility may be helpful for such problems resolving. --Oleg Stephen J. Gowdy wrote: There was nothing there. It did look like something called mpath didn't find any partitions. It ran after dmraid. I wonder if it is getting confused due there being a RAID controller on the machine? I just submited the same question to rhel5-users mailing list. I'll see if there are any other answers there. On Mon, 19 Mar 2007, Oleg Sadov wrote: Stephen J. Gowdy wrote: Hi All, Sorry for the off topic question but thought I might get an answer quicker here. I'm trying to install RHEL5 on a machine (currently running RHEL3) it has trouble reading the partition table somehow. I can read it fine with fdisk after pushing ALT-F2 to get the shell up. Is there a limit in the number of partitions anaconda understands? Here is the partition table; Standard limit for PC-like partition table -- 16 partitions (with extended part-n). May be you can find some useful messages at ALT-F3 or ALT-F4? --Oleg Disk /dev/hde: 40.0 GB, 40020664320 bytes 255 heads, 63 sectors/track, 4865 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Device BootStart EndBlocks Id System /dev/hde1 * 133265041 83 Linux /dev/hde234 4865 388130405 Extended /dev/hde5 1107 4865 30194136 83 Linux /dev/hde6 845 1106 2104515 83 Linux /dev/hde734 582 4409779+ 83 Linux /dev/hde8 583 713 1052226 83 Linux /dev/hde9 714 844 1052226 82 Linux swap Originally it was out of order but I fixed that and it still has the problem. The message starts "partition table on device hda was unreadable" (this device is hda during the installation), it wants you to create a new partition table. I'd rather not as I want to keep some partitions after the reinstall. Any ideas? regards, Stephen. -- /--------+-\ |Stephen J. Gowdy, SLAC | CERN Office: 32-2-A22| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| |http://calendar.yahoo.com/gowdy | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/ -- /--------+-\ |Stephen J. Gowdy, SLAC | CERN Office: 32-2-A22| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| |http://calendar.yahoo.com/gowdy | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/ -- /--------+-\ |Stephen J. Gowdy, SLAC | CERN Office: 32-2-A22| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| |http://calendar.yahoo.com/gowdy | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/
Re: Specifying current directory as destination
On Wed, 28 Mar 2007, Troy Dawson wrote: The current directory is . (period in english) US English at least. It is a "full stop" in British English. Don't know about other variants. -- /+---------\ |Stephen J. Gowdy, SLAC | CERN Office: 32-2-A22| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| |http://calendar.yahoo.com/gowdy | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/
Re: 64-bit application
I've seen a similar behaviour, but not such a big swing. The P4s Xeons were more than 40% slower than the Opterons. The current generation of Xeons however do bettter by 10% over the Opterons. Of course you need to also weigh in the cost of the chips, but looking at the system as a whole that isn't as large an effect. On Wed, 26 Sep 2007, Miles O'Neal wrote: Troy Dawson said... |It really comes down to your application. If possible, try it on two |comparable CPU setups, one AMD and one Intel. I've seen some wildly lopsided |tests, try to at least give them the same amount of memory and the same disks. |Then run your application on it, and see which is faster. | |That's how I decided I like the Opteron. On my tests (recompiling rpm's) the |Opteron beat the Xeon. But I saw other people with the exact same setup, and |for them the Xeon beat the Opteron. It all came down to the application. And that can change over time. For years we bought only AMD-based systems, because most of our apps consistently ran better on them. Then we found a couple that were decidely better on INtel. When we tested equivalent servers for the last set of compute farm systems, Intel won hands down. -Miles -- /----+---------\ |Stephen J. Gowdy, SLAC | CERN Office: 32-2-A22| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| | | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/
Re: WIFI problem with ipw3945
I've found certain networks don't work well with this device. However, using NetworkManager makes it okay. Have you tried that? You need to turn on two services for it to work NetworkManager and NetworkManagerDispatcher. On Thu, 4 Oct 2007, Frederic Boone wrote: Hello, I have a Dell Latitude D620 with SL kernel 2.6.18-8.1.14.el5 installed. Everything works fine except the Wifi: Error message in system-config-network 1.3.99: "Définition des informations IP pour eth1. échoué. Aucun lien n'a é té trouvé. Vérifier le câble ?" (sorry this is french, it means "no link found. check cable?") I have installed all the ipw3945 related packages: ipw3945 ipw3945-firmware ipw3945d kernel-module-ipw3945-2.6.18-8.1.14.el5 The card seems to be dected (I also see the wifi led blinking): # /sbin/lsmod | grep ipw3945 ipw3945 178856 1 ieee80211 33417 1 ipw3945 but it seems to be unasociated: # /sbin/iwconfig eth1 eth1 unassociated ESSID:off/any Nickname:"lerma10.obspm.fr" Mode:Managed Frequency=nan kHz Access Point: Not-Associated Bit Rate:0 kb/s Tx-Power:16 dBm Retry limit:15 RTS thr:off Fragment thr:off Encryption key:off Power Management:off Link Quality:0 Signal level:0 Noise level:0 Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0 Tx excessive retries:0 Invalid misc:256 Missed beacon:0 I can however see the mac address in system-config-network. So I tried forcing this address with "iwconfig ap ..." but the result it the same... Any clue? Thanks -- /------------+-\ |Stephen J. Gowdy, SLAC | CERN Office: 32-2-A22| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| | | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/
Rejected posting to SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV (fwd)
Resend... --- Begin Message --- On the CERN network it is unstable but NetworkManager reconnects and my sessions just pause for a few moments. Not perfect but workable. I've no idea what makes the CERN network special. On Thu, 4 Oct 2007, Frederic Boone wrote: I've found certain networks don't work well with this device. However, using NetworkManager makes it okay. Have you tried that? You need to turn on two services for it to work NetworkManager and NetworkManagerDispatcher. Thanks for the tip. I run the two services but the connection is still unstable... I'll try on other wifi networks to see if this is related to this one. -- /+---------\ |Stephen J. Gowdy, SLAC | CERN Office: 32-2-A22| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| | | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/ --- End Message ---
Re: Upgrading offline?
Why is it harmful? It only updates already installed RPMs? Isn't -U and -i potentially more harmful? On Tue, 27 Nov 2007, Patrick Moelands wrote: 2007/11/27, John Summerfield <[EMAIL PROTECTED]>: No, you most certainly do not want -U if one's pointing at the whole repo, and if one follows my other advice, that's exactly the case. But using "-F" without care can also be harmful to your system. You wouldn't want to do that without the proper care ;-) Regards, Patrick -- /----+-----\ |Stephen J. Gowdy, SLAC | CERN Office: 32-2-A22| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| | | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/
Re: I wish your R RPM did not have the Epoch set
Ah, that is probably also why RHEL5's openoffice.org-2.0.x keeps wanting to install even though I have 2.3! On Sat, 1 Dec 2007, Paul Johnson wrote: I built R-2.6.0 from the fedora distribution and was using it in my SL5 systems. Today I noticed that when yum ran, it wanted to update that version with the old one from SL5, 2.4 . That was weird. Why install an old one? After some checking, I learned that the Epoch had been set at 2 on the RPM from the SL 5 distribution. I've been knocking around in the RPM building community for about 10 years and I've never been persuaded the epoch serves a valid purpose, especially when it comes to differentiating R-2.4 from R-2.6 or whatnot. The Epoch only frustrates people who are trying to stay up to date. The Epoch is a frustrating value in this case because so 2:2.4.1 appears to be newer than 0:2.6.0. Just singing the complaining song... -- /+---------\ |Stephen J. Gowdy, SLAC | CERN Office: 32-2-A22| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| | | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/
Re: Installing multiple versions of libraries
You probably tried it but just to make sure... did you try to upgrade it? "rpm -Uhv ...". On Wed, 5 Dec 2007, Paul Johnson wrote: I have the cairo-1.2.4 that comes from SL5, and need to install the newer cairo in order to build poppler, which in turn is required for TeXLive. I don't have trouble building the new cairo, and planned to install that and leave the old one as well. But it doesn't work. # rpm -ivh cairo-1.4.10-2sl5.x86_64.rpm Preparing...### [100%] file /usr/lib64/libcairo.so.2 from install of cairo-1.4.10-2sl5 conflicts with file from package cairo-1.2.4-3.el5 See the links created by ldconfig: # ls -la libcairo* -rwxr-xr-x 1 root root 305056 Apr 3 2007 libcairojava-1.0.so -rwxr-xr-x 1 root root 54008 Apr 3 2007 libcairojni-1.0.so lrwxrwxrwx 1 root root 17 Dec 5 15:57 libcairo.so -> libcairo.so.2.9.2 lrwxrwxrwx 1 root root 17 Dec 5 15:57 libcairo.so.2 -> libcairo.so.2.9.2 -rwxr-xr-x 1 root root 431288 Nov 29 11:11 libcairo.so.2.9.2 It is a little bit confusing to me that the newer cairo still uses the old soname. Shouldn't it step up to libcairo.so.3 or something, in order to install both at same time? [EMAIL PROTECTED] x86_64]# rpm -qilp cairo-1.4.10-2sl5.x86_64.rpm Name: cairoRelocations: (not relocatable) Version : 1.4.10Vendor: (none) Release : 2sl5 Build Date: Wed 05 Dec 2007 02:27:06 PM CST Install Date: (not installed) Build Host: pols110.pols.ku.edu Group : System Environment/Libraries Source RPM: cairo-1.4.10-2sl5.src.rpm Size: 1544504 License: LGPL/MPL Signature : (none) URL : http://cairographics.org Summary : A vector graphics library Description : Cairo is a vector graphics library designed to provide high-quality display and print output. Currently supported output targets include the X Window System, OpenGL (via glitz), in-memory image buffers, and image files (PDF, PostScript, and SVG). Cairo is designed to produce identical output on all output media while taking advantage of display hardware acceleration when available (eg. through the X Render Extension or OpenGL). /usr/lib64/libcairo.so.2 /usr/lib64/libcairo.so.2.11.5 /usr/share/doc/cairo-1.4.10 /usr/share/doc/cairo-1.4.10/AUTHORS /usr/share/doc/cairo-1.4.10/BIBLIOGRAPHY /usr/share/doc/cairo-1.4.10/BUGS /usr/share/doc/cairo-1.4.10/COPYING /usr/share/doc/cairo-1.4.10/COPYING-LGPL-2.1 /usr/share/doc/cairo-1.4.10/COPYING-MPL-1.1 /usr/share/doc/cairo-1.4.10/ChangeLog /usr/share/doc/cairo-1.4.10/NEWS /usr/share/doc/cairo-1.4.10/PORTING_GUIDE /usr/share/doc/cairo-1.4.10/README /usr/share/doc/cairo-1.4.10/ROADMAP /usr/share/doc/cairo-1.4.10/TODO I can't just uninstall cairo-1.2.4, because it appears there are about 50 programs that are linked against it. What do you think? give up on the dream of replaciing tetex with TeXLive? Its kindof an overdue change, really. -- /----+-----\ |Stephen J. Gowdy, SLAC | CERN Office: 32-2-A22| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| | | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/
Re: USB key problem
If you write to the linux-usb mailing list they'll probably ask you to recompile the kernel with the USB verbose debugging on. I suspect that is what is needed to figure out this problem. I'd suspect there is some hardware problem that isn't worked around. If you are plugging it into the front USB socket on a machine you are using a USB_2_ rated cable internally in the machine, right? If there are other cables involved they are also rated for USB2? On Fri, 1 Aug 2008, Troy Dawson wrote: Michel Peru wrote: Hi all, The only way that I have found to make my USB key working with my PC runn ing SL5.2 is to do a "modprobe -r ehci_hcd". That works fine but unfortunately I can only transfer data at the rate of a USB1.1 device... My PC is based on a Asus K8N4E mother board with a Nvidia chipset. Any help to make my USB key working at a USB2.0 data rate would be greatl y appreciated. Michel Peru Sorry for the delay, but I thought someone else would have looked into this by now. A bit more information is needed. Can you send us the following lspci and cat /etc/modprobe.conf Thanks Troy -- /+-------------\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://www.slac.stanford.edu/~gowdy/ | CH-1211 Geneva 23| | | Switzerland | |EMail: [EMAIL PROTECTED] | Tel: +41 22 767 5840 | \+-/
Re: samba on SL5
Hi Devin, Wouldn't it be best to run SAMBA on the machines that actually have the disks? regards, Stephen. On Tue, 3 Feb 2009, Devin Bougie wrote: On Feb 3, 2009, at 4:10 PM, Dr Andrew C Aitchison wrote: On Tue, 3 Feb 2009, Devin Bougie wrote: Hi, All. We are using samba on an SL5.2 server to share filesystems with our Windows systems. Everything works properly when sharing a filesystem that is local to the samba server. When sharing an nfs filesystem, however, we see locking issues with *some* file types. Which kernel are you using on the *NFS* server ? We had NFS locking issues with the kernel shipped with SL5.2 and some of the updates. We don't seem to get them with 2.6.18-92.1.22.el5 It may be that your macs don't see the problem because they are being less picky with the locks. We see the same problem with all of the NFS shares we've tried, and none of the NFS servers are running SL5. One is Solaris 10, one is SL4 with the 2.6.9-67.0.1.ELsmp kernel, and several are SL3 with the 2.4.21-37.EL.XFSsmp kernel. If anyone else is serving nfs shares over samba on SL5 and isn't seeing this problem, I would be very grateful for a glimpse of your smb.conf file. If anyone is curious, I would be happy to provide ours. Many thanks for the reply, Devin For us the broken locking made mutt very unhappy with an NFS mounted /var/spool/mail but we found no problems with pine. -- Dr. Andrew C. Aitchison Computer Officer, DPMMS, Cambridge a.c.aitchi...@dpmms.cam.ac.uk http://www.dpmms.cam.ac.uk/~werdna -- /+-------------\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/
Re: SL 5.2 NetworkManager* problem
Hi Jon, I think this was fixed yesterday. I had the same issue but my update overnight worked. You're mirror probably just needs updated. regards, Stephen. On Fri, 27 Mar 2009, Jon Brinkmann wrote: The output from "yum update" on or SL 5.2 computers just started saying: Error: Missing Dependency: dbus >= 1.1 is needed by package NetworkManager Error: Missing Dependency: dbus-glib >= 0.73-6 is needed by package NetworkManager-glib Error: Missing Dependency: dbus-glib >= 0.73-6 is needed by package NetworkManager Error: Missing Dependency: dbus-glib >= 0.73-6 is needed by package NetworkManager-gnome Error: Missing Dependency: dbus >= 1.1 is needed by package NetworkManager-glib Error: Missing Dependency: dbus >= 1.1 is needed by package NetworkManager-gnome The versions installed are: # yum list dbus dbus-glib Loading "kernel-module" plugin Installed Packages dbus.i3861.0.0-7.el5_2.1 installed dbus-glib.i386 0.70-5 installed Attempting to update dbus and dbug gives: # yum update dbus dbus-glib Loading "kernel-module" plugin Setting up Update Process Could not find update match for dbus Could not find update match for dbus-glib No Packages marked for Update It looks like NetworkManager* was updated with a newer version of dbus and dbus-glib than is available in SL 5.2. Please correct the problem. Thanks, Jon --- Dr. Jon Brinkmann, Factotum Apache Point Observatory Network and Systems Administrator 2001 Apache Point Road Instrument ScientistP.O. Box 59 E-mail: brinkm...@nmsu.edu Sunspot, NM 88349-0059 URL:http://www.apo.nmsu.edu/brinkmann/ Tel: 575-437-6822 -- /--------+-\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/
Re: Problems using X Windows Display
some complaints, but fails on others, complaining about being unable to open display. (Unfortunately I cannot reproduce the failure on this machine, in what is given below.) Xnest does not seem to permit it. How can I go about this? Example output: [...@hpsl5 ~]$ Xnest :1 [1]+ Stopped Xnest :1 [...@hpsl5 ~]$ bg [1]+ Xnest :1 & [...@hpsl5 ~]$ xterm -display :1 AUDIT: Mon May 4 01:59:19 2009: 27279 Xnest: client 1 rejected from local host Xlib: connection to ":1.0" refused by server Xlib: No protocol specified xterm Xt error: Can't open display: :1 [...@hpsl5 ~]$ su Password: [r...@hpsl5 wss]# xterm -display :1 AUDIT: Mon May 4 02:00:23 2009: 27279 Xnest: client 1 rejected from local host Xlib: connection to ":1.0" refused by server Xlib: No protocol specified Warning: This program is an suid-root program or is being run by the root user. The full text of the error or warning message cannot be safely formatted in this environment. You may get a more descriptive message by running the program as a non-root user or by removing the suid bit on the executable. xterm Xt error: Can't open display: %s [r...@hpsl5 wss]# exit exit [...@hpsl5 ~]$ su wsshu Password: [ws...@hpsl5 wss]$ xterm -display :1 AUDIT: Mon May 4 02:01:44 2009: 27279 Xnest: client 1 rejected from local host Xlib: connection to ":1.0" refused by server Xlib: No protocol specified xterm Xt error: Can't open display: :1 [ws...@hpsl5 wss]$ cd [ws...@hpsl5 ~]$ xterm -display :1 AUDIT: Mon May 4 02:02:33 2009: 27279 Xnest: client 1 rejected from local host Xlib: connection to ":1.0" refused by server Xlib: No protocol specified xterm Xt error: Can't open display: :1 [ws...@hpsl5 ~]$ Xnest :2 & [1] 27357 [ws...@hpsl5 ~]$ xterm -display :2 AUDIT: Mon May 4 02:03:07 2009: 27357 Xnest: client 1 rejected from local host Xlib: connection to ":2.0" refused by server Xlib: No protocol specified xterm Xt error: Can't open display: :2 [ws...@hpsl5 ~]$ evince icegov2008-registration-noCardDetails-wss.pdf & [2] 27386 [ws...@hpsl5 ~]$ (evince:27386): GnomeUI-WARNING **: While connecting to session manager: Authentication Rejected, reason : None of the authentication protocols specified are supported and host-based authentication failed. ** (evince:27386): WARNING **: Service registration failed. ** (evince:27386): WARNING **: Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken. [2]+ Done evince icegov2008-registration-noCardDetails-wss.pdf [ws...@hpsl5 ~]$ William. -- /+-\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/
Re: Problems using X Windows Display
you don't get xauth. To check (on SL5) just do rpm -q xorg-x11-xauth Also, for me, when I check to see what my display setting is, I always do echo $DISPLAY And it should come back something like localhost:10.0 That is because it's doing an ssh tunnel, so it thinks it's the localhost. Hope this helps Troy Stephen J. Gowdy wrote: Hi William, X displays usually are setup to enforce some sort of security. Otherwise anyone would be able to read your password. In case 1, was DISPLAY set on hpsl5 before you typed ssh? I'm not sure case 2 is possible. You should use some sort of conferencing system to allow remote users to see your display (like EVO). For case 3 whoever is logged in the X-window should be allowed to open windows. xauth is used normally to manage authorisation and you could enable others to open windows on the local machine by extracting the correct key from whoever has started the x-windows session. If I assume it is wss, he would type something like; xauth list > auth.list then wsshu would type; xauth merge ~wss/auth.list assuming he is able to read that file. If not you should copy it somewhere wsshu can read it. This probably only works till wss exits his X session. Remember though, you are giving everything you type or see that other user. regards, Stephen. On Sun, 3 May 2009, William Shu wrote: I am having difficulties related to X Windowing system. Being a novice and basically overwhelmed by the X.org documentation. I present below the separate but related issues which can be summarised as: (1) displaying files from remote machines, possibly over secure shell; (2) projecting a window or entire screen onto *multiple* remote displays (monitors); and (3) using the same display when logged on as distinct users in xterm windows. In the example, I am on the host machine is hpsl5 (IP: 192.168.10.4) running SL5.0 and the remote machine is inteksl52 (192.168.10.20) running SL 5.2. Any assistance would be appreciated. ISSUE 1 = Displaying *.pdf *.ps files from a remote machine using secure shell ssh -XY, whereas it used to work (in the distant past). I get the message: "Unable to open the diplay" . Even trying to open a specific display (192.168.10.4:0, on tinysl5) with the xlsfonts command, I still get the message "Unable to open the diplay". Unfortunately, I do not have a very clear idea how X works; the manpage X(7) is not too helpful, and dmesg does not issue any messages. Example output: [...@hpsl5 ~]$ ssh -XY w...@192.168.10.20 w...@192.168.10.20's password: Last login: Fri May 1 00:45:23 2009 from 192.168.10.4 [...@inteksl52 ~]$ printenv |grep -ie display [...@inteksl52 ~]$ dir *.ps tsi.comp.POST.SENT-13032007_pages25_26.ps [...@inteksl52 ~]$ gv tsi.comp.POST.SENT-13032007_pages25_26.ps & [1] 23151 [...@inteksl52 ~]$ gv: Unable to open the display. [...@inteksl52 ~]$ [...@inteksl52 ~]$ xlsfonts -fn '-*-*-*-*-*-*-0-0-0-0-*-0-*-*' xlsfonts: unable to open display '' usage: xlsfonts [-options] [-fn pattern] where options include: -l[l[l]] give long info about each font -m give character min and max bounds -C force columns -1 force single column -u keep output unsorted -o use OpenFont/QueryFont instead of ListFonts -w width maximum width for multiple columns -n columns number of columns if multi column -display displayname X server to contact -d displayname (alias for -display displayname) [...@inteksl52 ~]$ [...@inteksl52 ~]$ [...@inteksl52 ~]$ xlsfonts -d 192.168.10.4:0.0 -fn '-*-*-*-*-*-*-0-0-0-0-*-0-*-*' xlsfonts: unable to open display '192.168.10.4:0.0' usage: xlsfonts [-options] [-fn pattern] where options include: -l[l[l]] give long info about each font -m give character min and max bounds -C force columns -1 force single column -u keep output unsorted -o use OpenFont/QueryFont instead of ListFonts -w width maximum width for multiple columns -n columns number of columns if multi column -display displayname X server to contact -d displayname (alias for -display displayname) ISSUE 2 = How can I display a given window (xterm, pdf file, etc.) on a number of remote terminal? For exmple, I would want that the pdf file I am scrolling through is also visible to my remote audi
Re: Problems using X Windows Display
On Mon, 4 May 2009, William Shu wrote: Thank you very much Miles! I rebooted both machines and I can now view pdf/ *.ps files without complaints! vncviewer now opens on the host machine when invoked from the remote machine through ssh. However, I have the following three problems (whose solutions I guess will help end this thread): 1) Everything has suddenly become so slow on the host machine (I think); shortly after opening the vnc window seems to freeze, not displaying X clients, though the window itself can be reduced or expanded! Sorry, no idea there. 2) Trying to open vnc without ssh complains of no route to host (error 113). Also, xclock complains of inability to open display. This is probably your firewall preventing access as I mentinoed. Looking at it I think you need to open 5901 on your server. I've never used though so I'm not sure. 3) Related to (2) above, it seems I can only project X disply on another terminal through ssh, which requires login (and hence knowing another user's password). Is there no other secure way of doing this, where the target user can selectively authorise the display? Each user could have an account on the host machine that is different. They'd all need to know the shared password for the VNC server though. Probably not a good idea. You can use xauth as described before to allow a user to do everything with your X session either via an ssh tunnel or directly if you open port 6000 (assuming your display is :0) on your firewall. However, something like EVO would be a much safer way to share a desktop window. There are probably other applications that would do it too. The following outputs show what transpired: [...@hpsl5 ~]$ ssh -XY w...@192.168.10.20 w...@192.168.10.20's password: Last login: Tue May 5 02:57:19 2009 [...@inteksl52 ~]$ echo $DISPLAY localhost:10.0 [1]+ Done evince Desktop/semanticKnowledge-a4-geissler.pdf [...@inteksl52 ~]$ vncviewer & [1] 6612 [...@inteksl52 ~]$ VNC Viewer Free Edition 4.1.2 for X - built Feb 11 2009 12:55:24 Copyright (C) 2002-2005 RealVNC Ltd. See http://www.realvnc.com for information on VNC. Tue May 5 03:08:35 2009 CConn: connected to host localhost port 5901 CConnection: Server supports RFB protocol version 3.8 CConnection: Using RFB protocol version 3.8 Tue May 5 03:08:54 2009 TXImage: Using default colormap and visual, TrueColor, depth 24. CConn: Using pixel format depth 6 (8bpp) rgb222 CConn: Using ZRLE encoding Tue May 5 03:18:27 2009 main: End of stream [1]+ Done vncviewer [...@inteksl52 ~]$ <... snip ...> [...@hpsl5 ~]$ vncviewer 192.168.10.20:1 & [1] 3888 [...@hpsl5 ~]$ VNC Viewer Free Edition 4.1.2 for X - built Feb 11 2009 12:55:24 Copyright (C) 2002-2005 RealVNC Ltd. See http://www.realvnc.com for information on VNC. Tue May 5 03:38:32 2009 main: unable to connect to host: No route to host (113) [1]+ Exit 1 vncviewer 192.168.10.20:1 [...@hpsl5 ~]$ [...@hpsl5 ~]$ xclock -display 192.168.10.20:1 & [1] 5564 [...@hpsl5 ~]$ Error: Can't open display: 192.168.10.20:1 --- On Tue, 5/5/09, Miles O'Neal wrote: From: Miles O'Neal Subject: Re: Problems using X Windows Display To: ws...@yahoo.com Date: Tuesday, May 5, 2009, 1:02 AM William Shu said... |Xforwarding: | |Changed the Xforwarding to yes in /etc/ssh/sshd_config of remote mach= |ine (inteksl52): | |#X11Forwarding no |X11Forwarding yes Did you restart the ssh daemon after that change? -- /------------+-\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/
Re: Problems using X Windows Display
vic can probably do it if you use multicast locally; http://mediatools.cs.ucl.ac.uk/nets/mmedia/ but that is probably whole new kettle of fish. On Tue, 5 May 2009, William Shu wrote: Dear All, I think I now have a clearer approach to finding solutions to some of my problems. the suggestions of Stephen, Troy and Miles were particularly helpful. First, the the solutions obtained so far: enabling Xforwarding and restarting daemons (e.g., sshd) permits me to have trouble-free displays over ssh or from xterm windows of different users on the same console. From the help offered, I think my solution strategy for multi-terminal display could be one of the following: 1) Bring up vnc display, have a "master vnc viewer" that can read/write on the display, and let all the other vnc viewers be "slave vnc viewers" that can only view the display. Unfortunately, blocking keyboard/mouse actions using the options menu from pressing the F8 key can be reset by the user. An suggestions to configure vnc to have such master-slave viewers? 2) Create a separate account, which can possibly become insecure by granting xauth authorities, and use vnc to display whatever. While the implications of extending access (via xauth) frighten, the use of a video conferencing tool, such as EVO, seem to require internet access, whereas the machines I'll use are in a closed LAN. Does anyone know of a video conferencing tool that does not register to/through the internet? Once more thank you all for the support. William. --- On Tue, 5/5/09, Stephen J. Gowdy wrote: From: Stephen J. Gowdy Subject: Re: Problems using X Windows Display To: "William Shu" Cc: "Miles O'Neal" , scientific-linux-us...@fnal.gov Date: Tuesday, May 5, 2009, 7:44 AM On Mon, 4 May 2009, William Shu wrote: Thank you very much Miles! I rebooted both machines and I can now view pdf/ *.ps files without complaints! vncviewer now opens on the host machine when invoked from the remote machine through ssh. However, I have the following three problems (whose solutions I guess will help end this thread): 1) Everything has suddenly become so slow on the host machine (I think); shortly after opening the vnc window seems to freeze, not displaying X clients, though the window itself can be reduced or expanded! Sorry, no idea there. 2) Trying to open vnc without ssh complains of no route to host (error 113). Also, xclock complains of inability to open display. This is probably your firewall preventing access as I mentinoed. Looking at it I think you need to open 5901 on your server. I've never used though so I'm not sure. 3) Related to (2) above, it seems I can only project X disply on another terminal through ssh, which requires login (and hence knowing another user's password). Is there no other secure way of doing this, where the target user can selectively authorise the display? Each user could have an account on the host machine that is different. They'd all need to know the shared password for the VNC server though. Probably not a good idea. You can use xauth as described before to allow a user to do everything with your X session either via an ssh tunnel or directly if you open port 6000 (assuming your display is :0) on your firewall. However, something like EVO would be a much safer way to share a desktop window. There are probably other applications that would do it too. The following outputs show what transpired: [...@hpsl5 ~]$ ssh -XY w...@192.168.10.20 w...@192.168.10.20's password: Last login: Tue May 5 02:57:19 2009 [...@inteksl52 ~]$ echo $DISPLAY localhost:10.0 [1]+ Done evince Desktop/semanticKnowledge-a4-geissler.pdf [...@inteksl52 ~]$ vncviewer & [1] 6612 [...@inteksl52 ~]$ VNC Viewer Free Edition 4.1.2 for X - built Feb 11 2009 12:55:24 Copyright (C) 2002-2005 RealVNC Ltd. See http://www.realvnc.com for information on VNC. Tue May 5 03:08:35 2009 CConn: connected to host localhost port 5901 CConnection: Server supports RFB protocol version 3.8 CConnection: Using RFB protocol version 3.8 Tue May 5 03:08:54 2009 TXImage: Using default colormap and visual, TrueColor, depth 24. CConn: Using pixel format depth 6 (8bpp) rgb222 CConn: Using ZRLE encoding Tue May 5 03:18:27 2009 main: End of stream [1]+ Done vncviewer [...@inteksl52 ~]$ <... snip ...> [...@hpsl5 ~]$ vncviewer 192.168.10.20:1 & [1] 3888 [...@hpsl5 ~]$ VNC Viewer Free Edition 4.1.2 for X - built Feb 11 2009 12:55:24 Copyright (C) 2002-2005 RealVNC Ltd. See http://www.realvnc.com for information on VNC. Tue May 5 03:38:32 2009 main: unable to connect to host: No route to host (113) [1]+ Exit 1 vncviewer 192.168.10.20:1 [...@hpsl5 ~]$ [...@hpsl5 ~]$ xclock -display 192.168.10.20:1 & [1] 5564 [...@hpsl5 ~]$ Error: Can't open display: 192.168.10.20:1
Re: SL-5.1Need help to know name of rpms
google points at FairRoot (CBM version I'd assume from the names). On Mon, 7 Sep 2009, Matthias Schroeder wrote: Sangamesh B wrote: Dear SL-5.1 users, I'm trying to install an application on CentsOS-5.2. The application works well on ScientificLinux-5.1but its failing to run on CentsOS-5.2(compilation goes smooth). That's because CentOS is lacking some of the rpms compared to SL-5.1. As the program fails by just showing .so not found error, I'm not able to find from which rpm that library come from. Following is the list of missing libraries: libGeoBase.so libParBase.so libBase.so libCbmBase.so libCbmData.so libField.so libGen.so libPassive.so libSts.so libTrd.so libTof.so libMuch.so To me this sounds like root libs or experiment specific libs, but nothing coming from SL, whatever release. Matthias As I don't have SL-5.1, I request any one of you to check which rpm these library belong to, i.e. the following 2 commands $ locate libGeoBase It will give the path of libGeoBase $ rpm -qf It will give the name of rpm. Same steps should be carried for other libraries also. Send me the list of rpms. Thanks for your kind help -- /+-------------\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/
Re: SL-5.1Need help to know name of rpms
I think you need to talk to whoever gave you the application. On Mon, 7 Sep 2009, Sangamesh B wrote: Thanks Stephen and Matthias for your quick update. Stephen, I also had did a googling and found about FairRoot, Cbmroot and something like Panda also. But didn't get how/which-one to install that would give the required libraries and resolve the issue. On Mon, Sep 7, 2009 at 1:00 PM, Stephen J. Gowdy wrote: google points at FairRoot (CBM version I'd assume from the names). Yes. On Mon, 7 Sep 2009, Matthias Schroeder wrote: Sangamesh B wrote: Dear SL-5.1 users, I'm trying to install an application on CentsOS-5.2. The application works well on ScientificLinux-5.1but its failing to run on CentsOS-5.2(compilation goes smooth). That's because CentOS is lacking some of the rpms compared to SL-5.1. As the program fails by just showing .so not found error, I'm not able to find from which rpm that library come from. Following is the list of missing libraries: libGeoBase.so libParBase.so libBase.so libCbmBase.so libCbmData.so libField.so libGen.so libPassive.so libSts.so libTrd.so libTof.so libMuch.so To me this sounds like root libs or experiment specific libs, but nothing coming from SL, whatever release. Yes. You are right. While running the application/benchmark, I use root -l much_sim.C. How the root libs can be installed? Matthias As I don't have SL-5.1, I request any one of you to check which rpm these library belong to, i.e. the following 2 commands $ locate libGeoBase It will give the path of libGeoBase $ rpm -qf It will give the name of rpm. Same steps should be carried for other libraries also. Send me the list of rpms. Thanks for your kind help -- /----+---------\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/ -- /----+---------\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/
Re: SL-5.1Need help to know name of rpms
To run the Cbm simulation you need the Cbm software. You should ask them how you get it. It isn't part of SL. This wouldn't work on a SL-5.1 install without the Cbm software also being installed. On Mon, 7 Sep 2009, Sangamesh B wrote: But the enduser tells blindly, it works on SL-5.1. The error is as follows: # root -l much_sim.C root [0] Processing much_sim.C... CBMROOT Macro much_sim = First input file is /opt/bench-intel/cbmroot/macro/much/data/jpsi.root Second input file is /opt/bench-intel/cbmroot/auau/25gev/centr/urqmd.auau.25gev.centr..ftn14 Output file is data/mc.moduleoff.jpsi.1000.root Events to process: 1000 === === much_sim.C : Loading libraries ... PSaid instance created... access via gSaid->f() Error in : libGeoBase[.so | .sl | .dl | .a | .dll] does not exist in .:/opt/bench-intel/fairsoft/tools/root_v5.20.00/lib::/opt/bench-intel/fairsoft/tools/root_v5.20.00/lib:/opt/bench-intel/fairsoft/generators/lib:/opt/bench-intel/fairsoft/tools/root_v5.20.00/lib:/opt/n1ge62/lib/lx24-amd64:/opt/n1ge62/lib/lx24-amd64:/usr/local/src/ns-allinone-2.33/otcl-1.13:/usr/local/src/ns-allinone-2.33/lib:/opt/bench-intel/fairsoft/generators/lib:/opt/bench-intel/fairsoft/tools/root_v5.20.00/lib:/opt/n1ge62/lib/lx24-amd64:/opt/n1ge62/lib/lx24-amd64:/usr/local/src/ns-allinone-2.33/otcl-1.13:/usr/local/src/ns-allinone-2.33/lib::/opt/bench-intel/fairsoft/tools/root_v5.20.00/cint/cint/stl Error in : libParBase[.so | .sl | .dl | .a | .dll] does not exist in .:/opt/bench-intel/fairsoft/tools/root_v5.20.00/lib::/opt/bench-intel/fairsoft/tools/root_v5.20.00/lib:/opt/bench-intel/fairsoft/generators/lib:/opt/bench-intel/fairsoft/tools/root_v5.20.00/lib:/opt/n1ge62/lib/lx24-amd64:/opt/n1ge62/lib/lx24-amd64:/usr/local/src/ns-allinone-2.33/otcl-1.13:/usr/local/src/ns-allinone-2.33/lib:/opt/bench-intel/fairsoft/generators/lib:/opt/bench-intel/fairsoft/tools/root_v5.20.00/lib:/opt/n1ge62/lib/lx24-amd64:/opt/n1ge62/lib/lx24-amd64:/usr/local/src/ns-allinone-2.33/otcl-1.13:/usr/local/src/ns-allinone-2.33/lib::/opt/bench-intel/fairsoft/tools/root_v5.20.00/cint/cint/stl I guess this could be a C++ package. Do you get any hints? Thanks, Sangamesh On Mon, Sep 7, 2009 at 1:35 PM, Stephen J. Gowdy wrote: I think you need to talk to whoever gave you the application. On Mon, 7 Sep 2009, Sangamesh B wrote: Thanks Stephen and Matthias for your quick update. Stephen, I also had did a googling and found about FairRoot, Cbmroot and something like Panda also. But didn't get how/which-one to install that would give the required libraries and resolve the issue. On Mon, Sep 7, 2009 at 1:00 PM, Stephen J. Gowdy wrote: google points at FairRoot (CBM version I'd assume from the names). Yes. On Mon, 7 Sep 2009, Matthias Schroeder wrote: Sangamesh B wrote: Dear SL-5.1 users, I'm trying to install an application on CentsOS-5.2. The application works well on ScientificLinux-5.1but its failing to run on CentsOS-5.2(compilation goes smooth). That's because CentOS is lacking some of the rpms compared to SL-5.1. As the program fails by just showing .so not found error, I'm not able to find from which rpm that library come from. Following is the list of missing libraries: libGeoBase.so libParBase.so libBase.so libCbmBase.so libCbmData.so libField.so libGen.so libPassive.so libSts.so libTrd.so libTof.so libMuch.so To me this sounds like root libs or experiment specific libs, but nothing coming from SL, whatever release. Yes. You are right. While running the application/benchmark, I use root -l much_sim.C. How the root libs can be installed? Matthias As I don't have SL-5.1, I request any one of you to check which rpm these library belong to, i.e. the following 2 commands $ locate libGeoBase It will give the path of libGeoBase $ rpm -qf It will give the name of rpm. Same steps should be carried for other libraries also. Send me the list of rpms. Thanks for your kind help -- /------------+-\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/ -- /------------+-\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/ -- /------------+-\ |
Re: problem mounting mass storage of mobile phone via usb on SL51
usb-storage module should be loaded automatically and it should end up in /proc/scsi/scsi. I have the same phone so could have checked if I had the USB cable with me... I have used it with Fedora 12 after selecting Storage Mode on the phone (it looks like you did from below). What does /proc/bus/usb/devices show? It should have usb-storage listed as the Driver= for it. On Thu, 4 Feb 2010, Gasser Marc wrote: Hello, anybody has an idea how to mount a mobile phone via usb in SL51. Neither fdisk -l nor parted can see a device file for the mobile, only lsusb (see below) shows that it is plugged in. Thanks for help. Regards, Marc Gasser PSI-Villigen, CH ### output of lsusb -v Bus 007 Device 004: ID 0421:01c7 Nokia Mobile Phones Device Descriptor: bLength18 bDescriptorType 1 bcdUSB 2.00 bDeviceClass0 (Defined at Interface level) bDeviceSubClass 0 bDeviceProtocol 0 bMaxPacketSize064 idVendor 0x0421 Nokia Mobile Phones idProduct 0x01c7 bcdDevice3.16 iManufacturer 1 Nokia iProduct2 N900 (Storage Mode) iSerial 3 372041756775 bNumConfigurations 2 Configuration Descriptor: bLength 9 bDescriptorType 2 wTotalLength 32 bNumInterfaces 1 bConfigurationValue 1 iConfiguration 4 Max power bmAttributes 0x80 MaxPower 500mA Interface Descriptor: bLength 9 bDescriptorType 4 bInterfaceNumber0 bAlternateSetting 0 bNumEndpoints 2 bInterfaceClass 8 Mass Storage bInterfaceSubClass 6 SCSI bInterfaceProtocol 80 Bulk (Zip) iInterface 6 Mass Storage Endpoint Descriptor: bLength 7 bDescriptorType 5 bEndpointAddress 0x81 EP 1 IN bmAttributes2 Transfer TypeBulk Synch Type None Usage Type Data wMaxPacketSize 0x0200 1x 512 bytes bInterval 0 Endpoint Descriptor: bLength 7 bDescriptorType 5 bEndpointAddress 0x01 EP 1 OUT bmAttributes2 Transfer TypeBulk Synch Type None Usage Type Data wMaxPacketSize 0x0200 1x 512 bytes bInterval 1 Configuration Descriptor: bLength 9 bDescriptorType 2 wTotalLength 32 bNumInterfaces 1 bConfigurationValue 2 iConfiguration 5 Self-powered bmAttributes 0xc0 Self Powered MaxPower 100mA Interface Descriptor: bLength 9 bDescriptorType 4 bInterfaceNumber0 bAlternateSetting 0 bNumEndpoints 2 bInterfaceClass 8 Mass Storage bInterfaceSubClass 6 SCSI bInterfaceProtocol 80 Bulk (Zip) iInterface 6 Mass Storage Endpoint Descriptor: bLength 7 bDescriptorType 5 bEndpointAddress 0x81 EP 1 IN bmAttributes2 Transfer TypeBulk Synch Type None Usage Type Data wMaxPacketSize 0x0200 1x 512 bytes bInterval 0 Endpoint Descriptor: bLength 7 bDescriptorType 5 bEndpointAddress 0x01 EP 1 OUT bmAttributes2 Transfer TypeBulk Synch Type None Usage Type Data wMaxPacketSize 0x0200 1x 512 bytes bInterval 1 Device Qualifier (for other device speed): bLength10 bDescriptorType 6 bcdUSB 2.00 bDeviceClass0 (Defined at Interface level) bDeviceSubClass 0 bDeviceProtocol 0 bMaxPacketSize064 bNumConfigurations 2 -- /+-\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/
Re: problem mounting mass storage of mobile phone via usb on SL51
Looks like it should work. Is there nothing in dmesg output about it? On Thu, 4 Feb 2010, Gasser Marc wrote: thank you for the reply. Well, it does and the driver is loaded (outputs below). Maybe you could test your phone once with SL and give me some feedback, that would be great. However, it's not urgent. Marc cat /proc/bus/usb/devices T: Bus=07 Lev=01 Prnt=01 Port=01 Cnt=01 Dev#= 4 Spd=480 MxCh= 0 D: Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs= 2 P: Vendor=0421 ProdID=01c7 Rev= 3.16 S: Manufacturer=Nokia S: Product=N900 (Storage Mode) S: SerialNumber=372041756775 C:* #Ifs= 1 Cfg#= 1 Atr=80 MxPwr=500mA I: If#= 0 Alt= 0 #EPs= 2 Cls=08(stor.) Sub=06 Prot=50 Driver=usb-storage E: Ad=81(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=01(O) Atr=02(Bulk) MxPS= 512 Ivl=125us C: #Ifs= 1 Cfg#= 2 Atr=c0 MxPwr=100mA I: If#= 0 Alt= 0 #EPs= 2 Cls=08(stor.) Sub=06 Prot=50 Driver= E: Ad=81(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=01(O) Atr=02(Bulk) MxPS= 512 Ivl=125us ### cat /proc/scsi/scsi Attached devices: Host: scsi0 Channel: 00 Id: 00 Lun: 00 Vendor: ATA Model: WDC WD2500AAJS-0 Rev: 01.0 Type: Direct-AccessANSI SCSI revision: 05 Host: scsi1 Channel: 00 Id: 00 Lun: 00 Vendor: TSSTcorp Model: CDDVDW TS-H653G Rev: FS00 Type: CD-ROM ANSI SCSI revision: 05 Host: scsi8 Channel: 00 Id: 00 Lun: 00 Vendor: NokiaModel: N900 Rev: 031 Type: Direct-AccessANSI SCSI revision: 02 Host: scsi8 Channel: 00 Id: 00 Lun: 01 Vendor: NokiaModel: N900 Rev: 031 Type: Direct-AccessANSI SCSI revision: 02 #### Stephen J. Gowdy wrote: usb-storage module should be loaded automatically and it should end up in /proc/scsi/scsi. I have the same phone so could have checked if I had the USB cable with me... I have used it with Fedora 12 after selecting Storage Mode on the phone (it looks like you did from below). What does /proc/bus/usb/devices show? It should have usb-storage listed as the Driver= for it. On Thu, 4 Feb 2010, Gasser Marc wrote: Hello, anybody has an idea how to mount a mobile phone via usb in SL51. Neither fdisk -l nor parted can see a device file for the mobile, only lsusb (see below) shows that it is plugged in. Thanks for help. Regards, Marc Gasser PSI-Villigen, CH ### output of lsusb -v Bus 007 Device 004: ID 0421:01c7 Nokia Mobile Phones Device Descriptor: bLength18 bDescriptorType 1 bcdUSB 2.00 bDeviceClass0 (Defined at Interface level) bDeviceSubClass 0 bDeviceProtocol 0 bMaxPacketSize064 idVendor 0x0421 Nokia Mobile Phones idProduct 0x01c7 bcdDevice3.16 iManufacturer 1 Nokia iProduct2 N900 (Storage Mode) iSerial 3 372041756775 bNumConfigurations 2 Configuration Descriptor: bLength 9 bDescriptorType 2 wTotalLength 32 bNumInterfaces 1 bConfigurationValue 1 iConfiguration 4 Max power bmAttributes 0x80 MaxPower 500mA Interface Descriptor: bLength 9 bDescriptorType 4 bInterfaceNumber0 bAlternateSetting 0 bNumEndpoints 2 bInterfaceClass 8 Mass Storage bInterfaceSubClass 6 SCSI bInterfaceProtocol 80 Bulk (Zip) iInterface 6 Mass Storage Endpoint Descriptor: bLength 7 bDescriptorType 5 bEndpointAddress 0x81 EP 1 IN bmAttributes2 Transfer TypeBulk Synch Type None Usage Type Data wMaxPacketSize 0x0200 1x 512 bytes bInterval 0 Endpoint Descriptor: bLength 7 bDescriptorType 5 bEndpointAddress 0x01 EP 1 OUT bmAttributes2 Transfer TypeBulk Synch Type None Usage Type Data wMaxPacketSize 0x0200 1x 512 bytes bInterval 1 Configuration Descriptor: bLength 9 bDescriptorType 2 wTotalLength 32 bNumInterfaces 1 bConfigurationValue 2 iConfiguration 5 Self-powered bmAttributes 0xc0 Self Powered MaxPower 100mA Interface Descriptor: bLength 9 bDescriptorType 4 bInterfaceNumber0 bAlternateSetting 0 bNumEndpoints 2 bInterfaceClass 8 Mass Storage bInterfaceSubClass 6 SCSI bInterfaceProtocol 80 Bulk (Zip) iInterface 6 Mass Storage
Re: cannot access ftp.scientificlinux.org
FNAL is mostly down at the moment. On Wed, 17 Feb 2010, Gasser Marc wrote: Hello, is the ftp server running? Regards, Marc -- /+-\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/
Re: Slow boot with Kernel 2.6.18-194.x.y on SL5.4
Remove "quiet" from /boot/grub/grub.conf. On Wed, 28 Jul 2010, Pom Sailasuta wrote: Can you please explain how to suppress 'quiet' during boot time? thanks Pom On Wed, Jul 28, 2010 at 4:38 AM, Alan Bartlett mailto:a...@elrepo.org>> wrote: On 28 July 2010 02:21, Franchisseur Robert mailto:robert.franchiss...@lmd.jussieu.fr>> wrote: On our Dell laptops E6400 or E6500 the boot process is very long. It takes ~ 40 seconds between: Redhat NASH and INIT: if we suppress 'quiet' we see that the problem is after the line : device-mapper : dm-raid45 : initialized V0.25941 Waiting for driver initialization Robert, This is a known issue, with a bug logged upstream [1] that has recently been "CLOSED WONTFIX". I don't have a solution to this but would like to comment that you are seeing twice the average delay of everyone else with whom I have discussed this issue. Typically it is an 18 - 20 second delay. Alan. [1] https://bugzilla.redhat.com/show_bug.cgi?id=499955 -- /--------+-\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/
Re: Scientific Linux CERN distribution.
http://linux.web.cern.ch/linux/scientific.shtml On Fri, 24 Sep 2010, Maximilian Eberl wrote: Seems like your have a CERN distribution. Sorry for the dumb question: what is the difference between SL from ftp.scientificlinux.org and a CERN distribution from glitesoft.cern.ch ? Thank You for a short explanation. Maximilian -- /+-\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/
Re: coreutils for 64 bit
It says it only copied 2.1GB. You are runnig a 64bit OS. You reinstalld the same coreutils package. You need to change the format of the package names from "rpm -qa" if you want to see the architecture ("man rpm" should help you figure out how). On Wed, 1 Feb 2012, Andrey Y Shevel wrote: Hi, I just paid attention that utility 'dd' uses just 2 GB even I use greater block size (BS). For example = [root@pcfarm-10 ~]# dd if=/dev/zero of=/mnt/sdb/TestFile-S1 bs=12GB count=1 0+1 records in 0+1 records out 2147479552 bytes (2.1 GB) copied, 15.8235 seconds, 136 MB/s BTW, [root@pcfarm-10 ~]# uname -a Linux pcfarm-10.pnpi.spb.ru 2.6.18-274.17.1.el5xen #1 SMP Tue Jan 10 16:41:16 EST 2012 x86_64 x86_64 x86_64 GNU/Linux [root@pcfarm-10 ~]# cat /etc/issue Scientific Linux SL release 5.7 (Boron) Kernel \r on an \m I decided to reinstall coreutils: [root@pcfarm-10 ~]# yum reinstall coreutils.x86_64 Failed to set locale, defaulting to C Loaded plugins: kernel-module Setting up Reinstall Process Resolving Dependencies --> Running transaction check ---> Package coreutils.x86_64 0:5.97-34.el5 set to be updated --> Finished Dependency Resolution Beginning Kernel Module Plugin Finished Kernel Module Plugin Dependencies Resolved === Package Arch Version Repository Size === Reinstalling: coreutilsx86_645.97-34.el5 sl-base 3.6 M Transaction Summary === Remove0 Package(s) Reinstall 1 Package(s) Downgrade 0 Package(s) Total download size: 3.6 M Is this ok [y/N]: y Downloading Packages: coreutils-5.97-34.el5.x86_64.rpm| 3.6 MB 00:05 Running rpm_check_debug Running Transaction Test Finished Transaction Test Transaction Test Succeeded Running Transaction Installing : coreutils 1/1 Installed: coreutils.x86_64 0:5.97-34.el5 Complete! = However after that I see [root@pcfarm-10 ~]# ls -l /bin/dd -rwxr-xr-x 1 root root 41464 Jul 26 2011 /bin/dd [root@pcfarm-10 ~]# rpm -q --file /bin/dd coreutils-5.97-34.el5 [root@pcfarm-10 ~]# rpm -qa | grep coreutils policycoreutils-1.33.12-14.8.el5 policycoreutils-newrole-1.33.12-14.8.el5 coreutils-5.97-34.el5 policycoreutils-gui-1.33.12-14.8.el5 i.e. no package with name coreutils.x86_64 I failed to find anything on the topic in scientific linux mailing list. Does somebody know about dd for 64 bit ? Many thanks in advance, Andrey -- /--------+-\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/
Re: coreutils for 64 bit
Exactly if you type "man rpm" it will show you how you get it to print the arch string (usually i686 or x86_64). Since you seem unabel to read a man page what you want to type is; rpm -qa --queryformat "%{NAME}-%{VERSION}.%{ARCH}\n" | grep coreutils (or miss out the VERSION if you want to see somethign similar to yum) On Wed, 1 Feb 2012, Andrey Y. Shevel wrote: Hi Stephen, thanks for the reply. I am not sure that I do understand you (sorry for my stupidity). I have === [root@pcfarm-10 ~]# yum list | grep coreutil Failed to set locale, defaulting to C coreutils.x86_64 5.97-34.el5 installed policycoreutils.x86_64 1.33.12-14.8.el5 installed policycoreutils-gui.x86_64 1.33.12-14.8.el5 installed policycoreutils-newrole.x86_64 1.33.12-14.8.el5 installed [root@pcfarm-10 ~]# rpm -q --file /bin/dd coreutils-5.97-34.el5 = Presumably all packages are appropriate (they have suffix x86_64) as shown by yum. At the same time rpm does show packages without above suffixes = [root@pcfarm-10 ~]# rpm -qa | grep coreutil policycoreutils-1.33.12-14.8.el5 policycoreutils-newrole-1.33.12-14.8.el5 coreutils-5.97-34.el5 policycoreutils-gui-1.33.12-14.8.el5 = On Wed, 1 Feb 2012, Stephen J. Gowdy wrote: Date: Wed, 1 Feb 2012 11:32:40 +0100 (CET) From: Stephen J. Gowdy To: Andrey Y Shevel Cc: SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV Subject: Re: coreutils for 64 bit It says it only copied 2.1GB. You are runnig a 64bit OS. You reinstalld the same coreutils package. You need to change the format of the package names from "rpm -qa" if you want to see the architecture ("man rpm" should help you figure out how). On Wed, 1 Feb 2012, Andrey Y Shevel wrote: Hi, I just paid attention that utility 'dd' uses just 2 GB even I use greater block size (BS). For example = [root@pcfarm-10 ~]# dd if=/dev/zero of=/mnt/sdb/TestFile-S1 bs=12GB count=1 0+1 records in 0+1 records out 2147479552 bytes (2.1 GB) copied, 15.8235 seconds, 136 MB/s BTW, [root@pcfarm-10 ~]# uname -a Linux pcfarm-10.pnpi.spb.ru 2.6.18-274.17.1.el5xen #1 SMP Tue Jan 10 16:41:16 EST 2012 x86_64 x86_64 x86_64 GNU/Linux [root@pcfarm-10 ~]# cat /etc/issue Scientific Linux SL release 5.7 (Boron) Kernel \r on an \m I decided to reinstall coreutils: [root@pcfarm-10 ~]# yum reinstall coreutils.x86_64 Failed to set locale, defaulting to C Loaded plugins: kernel-module Setting up Reinstall Process Resolving Dependencies --> Running transaction check ---> Package coreutils.x86_64 0:5.97-34.el5 set to be updated --> Finished Dependency Resolution Beginning Kernel Module Plugin Finished Kernel Module Plugin Dependencies Resolved === Package Arch Version Repository Size === Reinstalling: coreutilsx86_645.97-34.el5 sl-base 3.6 M Transaction Summary === Remove0 Package(s) Reinstall 1 Package(s) Downgrade 0 Package(s) Total download size: 3.6 M Is this ok [y/N]: y Downloading Packages: coreutils-5.97-34.el5.x86_64.rpm| 3.6 MB 00:05 Running rpm_check_debug Running Transaction Test Finished Transaction Test Transaction Test Succeeded Running Transaction Installing : coreutils 1/1 Installed: coreutils.x86_64 0:5.97-34.el5 Complete! = However after that I see [root@pcfarm-10 ~]# ls -l /bin/dd -rwxr-xr-x 1 root root 41464 Jul 26 2011 /bin/dd [root@pcfarm-10 ~]# rpm -q --file /bin/dd coreutils-5.97-34.el5 [root@pcfarm-10 ~]# rpm -qa | grep coreutils policycoreutils-1.33.12-14.8.el5 policycoreutils-newrole-1.33.12-14.8.el5 coreutils-5.97-34.el5 policycoreutils-gui-1.33.12-14.8.el5 i.e. no package with name coreutils.x86_64 I failed to find anything on the topic in scientific linux mailing list. Does somebody know about dd for 64 bit ? Many thanks in advance, Andrey -- /------------+-\ |Stephen J. Gowdy | CERN Office: 8-1-11| |http://cern.ch/gowdy/| CH-1211 Geneva 23| | | Switzerland | |EMail: go...@cern.ch | Tel: +41 76 487 2215 | \+-/
Re: coreutils for 64 bit
Hi Andrey, Why would you want a block size in GB? I don't know what the actual limit for dd itself is, although it does seem to be exactly 2GiB. regards, Stephen. On Thu, 2 Feb 2012, Andrey Y. Shevel wrote: Hi Stephen, thank you for your reply. == [root@pcfarm-10 ~]# rpm -qa --queryformat "%{NAME}-%{VERSION}.%{ARCH}\n" | grep coreutils policycoreutils-1.33.12.x86_64 policycoreutils-newrole-1.33.12.x86_64 coreutils-5.97.x86_64 policycoreutils-gui-1.33.12.x86_64 = And obviously [root@pcfarm-10 ~]# arch x86_64 === The result is prety same as I shown earlier. And the same I see at CERN === [lxplus427] /afs/cern.ch/user/s/shevel > dd if=/dev/zero of=/tmp/testx64 bs=3GB count=1 0+1 records in 0+1 records out 2147479552 bytes (2.1 GB) copied, 5.91242 seconds, 363 MB/s [lxplus427] /afs/cern.ch/user/s/shevel > rpm -q --file /bin/dd coreutils-5.97-34.el5 [lxplus427] /afs/cern.ch/user/s/shevel > rpm -qa --queryformat "%{NAME}-%{VERSION}.%{ARCH}\n" | grep coreutil policycoreutils-1.33.12.x86_64 coreutils-5.97.x86_64 policycoreutils-gui-1.33.12.x86_64 === As far as I understand the main question is "is there 64 bit dd version which can operate more then 2GB value for BS in SL anyway?" Any answer (yes or no) is good to know. Many thanks, Andrey On Wed, 1 Feb 2012, Stephen J. Gowdy wrote: Date: Wed, 1 Feb 2012 19:10:14 +0100 (CET) From: Stephen J. Gowdy To: Andrey Y. Shevel Cc: SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV Subject: Re: coreutils for 64 bit Exactly if you type "man rpm" it will show you how you get it to print the arch string (usually i686 or x86_64). Since you seem unabel to read a man page what you want to type is; rpm -qa --queryformat "%{NAME}-%{VERSION}.%{ARCH}\n" | grep coreutils (or miss out the VERSION if you want to see somethign similar to yum) On Wed, 1 Feb 2012, Andrey Y. Shevel wrote: Hi Stephen, thanks for the reply. I am not sure that I do understand you (sorry for my stupidity). I have === [root@pcfarm-10 ~]# yum list | grep coreutil Failed to set locale, defaulting to C coreutils.x86_64 5.97-34.el5 installed policycoreutils.x86_64 1.33.12-14.8.el5 installed policycoreutils-gui.x86_64 1.33.12-14.8.el5 installed policycoreutils-newrole.x86_64 1.33.12-14.8.el5 installed [root@pcfarm-10 ~]# rpm -q --file /bin/dd coreutils-5.97-34.el5 = Presumably all packages are appropriate (they have suffix x86_64) as shown by yum. At the same time rpm does show packages without above suffixes = [root@pcfarm-10 ~]# rpm -qa | grep coreutil policycoreutils-1.33.12-14.8.el5 policycoreutils-newrole-1.33.12-14.8.el5 coreutils-5.97-34.el5 policycoreutils-gui-1.33.12-14.8.el5 ===== On Wed, 1 Feb 2012, Stephen J. Gowdy wrote: > Date: Wed, 1 Feb 2012 11:32:40 +0100 (CET) > From: Stephen J. Gowdy > To: Andrey Y Shevel > Cc: SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV > Subject: Re: coreutils for 64 bit > > It says it only copied 2.1GB. You are runnig a 64bit OS. You reinstalld > the same coreutils package. You need to change the format of the package > names from "rpm -qa" if you want to see the architecture ("man rpm" > should help you figure out how). > > On Wed, 1 Feb 2012, Andrey Y Shevel wrote: > > > Hi, > > > > I just paid attention that utility 'dd' uses just 2 GB even I use > > greater > > block size (BS). For example > > > > = > > [root@pcfarm-10 ~]# dd if=/dev/zero of=/mnt/sdb/TestFile-S1 bs=12GB > > count=1 > > 0+1 records in > > 0+1 records out > > 2147479552 bytes (2.1 GB) copied, 15.8235 seconds, 136 MB/s > > > > > > BTW, > > > > [root@pcfarm-10 ~]# uname -a > > Linux pcfarm-10.pnpi.spb.ru 2.6.18-274.17.1.el5xen #1 SMP Tue Jan 10 > > 16:41:16 EST 2012 x86_64 x86_64 x86_64 GNU/Linux > > [root@pcfarm-10 ~]# cat /etc/issue > > Scientific Linux SL release 5.7 (Boron) > > Kernel \r on an \m > > > > > > > > > > I decided to reinstall coreutils: > > > > [root@pcfarm-10 ~]# yum reinstall coreutils.x86_64 > > Failed to set locale, defaulting to C > > Loaded plugins: kernel-module > > Setting up Reinstall Process > > Resolving Dependencies > > --> Running transaction check > > ---> Package coreutils.x86_64 0:5.97-34.el5 set t
Re: coreutils for 64 bit
Hi Chris, I understand using lager than 32kB block size can help the throughput but I'd doubt you'd get advantage with a 2GB block size over a 8MB block size for most devices. It may also be due to my laptop only having 4GB of RAM but it is much better to use 8MB rather than 2GB for my SSD drive; [root@antonia ~]# time dd if=/dev/sda of=/scratch/gowdy/test bs=8MB count=256 256+0 records in 256+0 records out 204800 bytes (2.0 GB) copied, 36.1101 s, 56.7 MB/s real0m36.125s user0m0.002s sys 0m2.420s root@antonia ~]# time dd if=/dev/sda of=/scratch/gowdy/test bs=2GB count=1 1+0 records in 1+0 records out 20 bytes (2.0 GB) copied, 56.1444 s, 35.6 MB/s real0m56.738s user0m0.001s sys 0m14.715s (oops, and I should have said 8M and 2G bs I guess). 2MB buffer isn't much slower; [root@antonia ~]# time dd if=/dev/sda of=/scratch/gowdy/test bs=2MB count=1024 1024+0 records in 1024+0 records out 204800 bytes (2.0 GB) copied, 38.4204 s, 53.3 MB/s real0m38.781s user0m0.004s sys 0m2.410s regards, Stephen. On Mon, 6 Feb 2012, Chris Schanzle wrote: It's a shame the original question didn't explain what and why he was trying to do something with these large blocks. Huge block sizes are useful if you have lots of ram and are copying very large files on the same set of spindles. This minimizes disk seeking caused by head repositioning for reads and writes and is vastly more efficient than say, "cp" which often uses at most 32 KB reads/writes and relies on the VM system to flush the writes (buffered by dirtying memory pages) pages as it deems appropriate (tunables in /proc/sys/vm/dirty*). Anyway, let's look at what system calls 'dd' does: $ strace dd if=/dev/zero of=/dev/shm/deleteme bs=12G count=1 ... open("/dev/shm/deleteme", O_WRONLY|O_CREAT|O_TRUNC, 0666) = 3 dup2(3, 1) = 1 close(3)= 0 mmap(NULL, 12884914176, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x2af98c7a read(0, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 12884901888) = 2147479552 write(1, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 2147479552) = 2147479552 close(0)= 0 close(1)= 0 ... (count=2 is also interesting) Things to notice: 1. strace shows dd is issuing a 12GB read from the input descriptor (/dev/zero) but is getting a 'short read' from the kernel of 2GB. Short reads are not an error. 2. The "count=" option in the dd man page specifies that it limits the number of INPUT blocks. So it writes what it read (2GB) and quits. So it seems to be working as designed, though perhaps not as you want. Adding 'iflag=fullblock' will cause dd to perform multiple reads to fill the input block size. mmap(NULL, 12884914176, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x2b2d8735e000 read(0, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 12884901888) = 2147479552 read(0, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 10737422336) = 2147479552 read(0, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 8589942784) = 2147479552 read(0, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 6442463232) = 2147479552 read(0, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 4294983680) = 2147479552 read(0, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 2147504128) = 2147479552 read(0, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 24576) = 24576 write(1, "", 12884901888) = 2147479552 write(1, "", 10737422336) = 2147479552 write(1, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 8589942784) = 2147479552 write(1, "", 6442463232)= 2147479552 write(1, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 4294983680) = 2147479552 Notice how the writes empty the input 2GB at a time. Of course, all this reading/writing goes through typical VM buffering, so you might want to consider direct i/o: iflag=direct and oflag=direct. Which begs the question: how to encourage the kernel to allow larger read/write file buffers? Couldn't find that answer easily. Anyone? -c On 02/02/2012 12:32 PM, Stephen J. Gowdy wrote: Hi Andrey, Why would you want a block size in GB? I don't know what the actual limit for dd itself is, although it does seem to be exactly 2GiB.
Re: coreutils for 64 bit
Hi Chris, When I read and write to the same disk the 2GB bs helps (a lot), but if I just write to a normal disk the 2GB block size doesn't. I get about 52MB/s with 2 or 8MB block size but only 44MB/s with a 2GB block size. This is just a standard disk, no RAID. regards, Stephen. On Mon, 6 Feb 2012, Chris Schanzle wrote: Hi Stephen, Most of my comments were in the context of reducing disk seeks. Using an SSD kinda eliminates that penalty. :-) It is unclear why your SSD writes big blocks slower than small blocks. SSD's are very complex little devices. Their write performance depends so much on the firmware's ability to have pre-erased, ready-to-write sectors (since erasing is slow), as well as writing on proper boundaries (like 'advanced format' 4k sector drives) to avoid read/modify/write cycles. TRIM/discard support is vital to maintaining performance over time. In one EL5 system, I have a md RAID0 (thus no TRIM) that has the *worst* random write performance of any system (including 'spinning rust' hard drives) now that it's aged and the firmware basically doesn't have any free pre-erased blocks. Thanks for showing your results. It's always good to test. Using similar dd commands to yours, on a traditional hard drive, I get about 37 MB/sec (with a lot of disk seeking noise) with 32K or 8M blocks; with bs=2G I get about 48 MB/s. Not as much difference as I would have expected, but in my case my output file might not have been very far from the beginning of the disk (hard to tell with LVM), so seeks might not have been very distant. There was essentially no disk seeking with bs=2G. Throw 'iflag=direct oflag=direct' with 32KB blocks and I drop to 30 MB/s. If you're reading/writing to different spindles, then you want reasonably small block sizes to increase parallelism between the reading and writing. I.e., you don't want to wait for a 2GB read to complete before starting a write. In that case, letting the VM system handle writing in the background in parallel works fine. Optimize your read/write sizes for your device. E.g., RAID devices typically have 64 KB to 256 KB stripes, so you want to be at least that big or some multiple thereof. Regards, Chris On 02/06/2012 12:47 PM, Stephen J. Gowdy wrote: Hi Chris, I understand using lager than 32kB block size can help the throughput but I'd doubt you'd get advantage with a 2GB block size over a 8MB block size for most devices. It may also be due to my laptop only having 4GB of RAM but it is much better to use 8MB rather than 2GB for my SSD drive; [root@antonia ~]# time dd if=/dev/sda of=/scratch/gowdy/test bs=8MB count=256 256+0 records in 256+0 records out 204800 bytes (2.0 GB) copied, 36.1101 s, 56.7 MB/s real0m36.125s user0m0.002s sys 0m2.420s root@antonia ~]# time dd if=/dev/sda of=/scratch/gowdy/test bs=2GB count=1 1+0 records in 1+0 records out 20 bytes (2.0 GB) copied, 56.1444 s, 35.6 MB/s real0m56.738s user0m0.001s sys 0m14.715s (oops, and I should have said 8M and 2G bs I guess). 2MB buffer isn't much slower; [root@antonia ~]# time dd if=/dev/sda of=/scratch/gowdy/test bs=2MB count=1024 1024+0 records in 1024+0 records out 204800 bytes (2.0 GB) copied, 38.4204 s, 53.3 MB/s real0m38.781s user0m0.004s sys 0m2.410s regards, Stephen. On Mon, 6 Feb 2012, Chris Schanzle wrote: It's a shame the original question didn't explain what and why he was trying to do something with these large blocks. Huge block sizes are useful if you have lots of ram and are copying very large files on the same set of spindles. This minimizes disk seeking caused by head repositioning for reads and writes and is vastly more efficient than say, "cp" which often uses at most 32 KB reads/writes and relies on the VM system to flush the writes (buffered by dirtying memory pages) pages as it deems appropriate (tunables in /proc/sys/vm/dirty*). Anyway, let's look at what system calls 'dd' does: $ strace dd if=/dev/zero of=/dev/shm/deleteme bs=12G count=1 ... open("/dev/shm/deleteme", O_WRONLY|O_CREAT|O_TRUNC, 0666) = 3 dup2(3, 1) = 1 close(3)= 0 mmap(NULL, 12884914176, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x2af98c7a read(0, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 12884901888) = 2147479552 write(1, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 2147479552) = 2147479552 close(0)= 0 close(1)= 0 ..