users March 2012

users@lists.fedoraproject.org

247 participants
305 discussions

Re: parallel bash scripts
by Mark LaPierre 27 Mar '12

27 Mar '12

I don't understand what you are trying to do. Do you want to run these commands in series? Run one, wait for it to complete, then start the other? If that's where you want to go then leave off the trailing ampersand but keep the redirection. The output of the first program will appear in the terminal due to the redirection. When the first program ends the second program will start. That one needs redirection of standard out to standard error too so that it's output will appear in the terminal as well. Solution with trailing ampersand is appropriate for endless loops. Solution without trailing ampersand is appropriate for serial operation. I put the loop counter in the demo program just to stop the thing without having to us ps and kill. Without the loop counter they will continue to print inter-twined messages forever. Does the program webservice_setup.sh start programs itself that you are trying to read the output from? Are they correctly redirected to send their standard output to standard error? On 03/27/2012 10:01 PM, bruce wrote: > Hi mark. > > the following is part of what i have tried... and what isn't working. > i changed the test shell scripts to no longer have infinite loops. > they simply execute the php app one time. > > however, it still doesn't work. > > in the initial shell script, it calls a php app that does some mysql > processes. the php app has a number of echo/print msgs to track what's > going on for testing.. when running as a background process the shell > script/php app isn't working. > > the cmdline sits.. when i select the "return" key.. the process is > diplayed as being stopped... > > if i remove the background "&" the shell script runs... > > #### -- not working > /apps/parseapp2/webservice_setup.sh 1>&2& > > #### -- works > /apps/parseapp2/webservice_setup.sh > > /apps/parseapp2/webservice_setup.sh > ------------------------------- > echo "starting the osu webservice test" > mysql -uroot1 -pfoo< /apps/parseapp2/setup_tblDb.sql > mysql -uroot1 -pfoo< /apps/parseapp2/webservice/webservice_tbls.sql > /apps/parseapp2/parse_tbl_setup.php > echo "started the osu webservice test" > > > thoughts??? > > thanks > > > > > > On Tue, Mar 27, 2012 at 9:37 PM, Mark LaPierre<marklapier(a)aol.com> wrote: >> Check out the redirection at the end of each command. 1>&2 redirects the >> standard out of your child command to the standard error which then appears >> in the parent shell. At the end the last& launches your command into a >> background shell and then moves on to launch the next command. The >> redirections don't care if the command ever terminates. >> >> The result is that both commands are launched and the parent shell >> terminates leaving the standard error attached to the terminal that the >> parent was launched in. >> >> >> On 03/27/2012 09:08 PM, bruce wrote: >>> >>> marklapier(a)aol.com >>> >>> hey mark.... >>> >>> what you have, appears to be pretty close to what i had... except my >>> tests never ended... the loops are infinite... >>> >>> can i do a fpaste and have you take a look at what i have? >>> >>> -btuce >>> >>> >>> On Tue, Mar 27, 2012 at 9:00 PM, Mark LaPierre<marklapier(a)aol.com> wrote: >>>> >>>> On 03/27/2012 05:25 PM, bruce wrote: >>>>> >>>>> >>>>> hi. >>>>> >>>>> got a couple of test bash scripts. >>>>> >>>>> dog.sh, cat.sh >>>>> each script runs the underlying php in an endless loop. >>>>> >>>>> I'm trying to figure out how to run the scripts in parallel, from the >>>>> same parent shell script. something like: >>>>> >>>>> test.sh >>>>> >>>>> >>>>> where dog.sh would be : >>>>> -------------------------------- >>>>> while true >>>>> do >>>>> pgrep dog >>>>> if [ $? -ne 0 ] >>>>> then >>>>> /dog.php >>>>> fi >>>>> sleep 5 >>>>> done >>>>> >>>>> my current tests, run dog.sh, which runs the dog.php ... but the test >>>>> never gets to run cat.sh >>>>> >>>>> thoughts/comments... >>>>> >>>>> thanks >>>> >>>> >>>> Hey Bruce, >>>> >>>> Do you mean to run these subprograms in parallel or in series? >>>> >>>> cat.sh >>>> #! /bin/bash >>>> >>>> CAT=0 >>>> until [ $CAT -eq 10 ] >>>> do >>>> echo "Inside a dog it's too dark to read. $CAT" >>>> CAT=$[$CAT + 1] >>>> sleep 2 >>>> done >>>> >>>> >>>> dog.sh >>>> #! /bin/bash >>>> >>>> DOG=0 >>>> until [ $DOG -eq 10 ] >>>> do >>>> echo "Next to a dog a book is man's best friend. $DOG" >>>> DOG=$[$DOG + 1] >>>> sleep 2 >>>> done >>>> >>>> >>>> test.sh >>>> #! /bin/sh >>>> >>>> /home/mlapier/test/dog.sh 1>&2& >>>> /home/mlapier/test/cat.sh 1>&2& >>>> >>>> >>>> [mlapier@mushroom test]$ ./test.sh >>>> [mlapier@mushroom test]$ Next to a dog a book is man's best friend. 0 >>>> Inside a dog it's too dark to read. 0 >>>> Next to a dog a book is man's best friend. 1 >>>> Inside a dog it's too dark to read. 1 >>>> Next to a dog a book is man's best friend. 2 >>>> Inside a dog it's too dark to read. 2 >>>> Next to a dog a book is man's best friend. 3 >>>> Inside a dog it's too dark to read. 3 >>>> Next to a dog a book is man's best friend. 4 >>>> Inside a dog it's too dark to read. 4 >>>> Next to a dog a book is man's best friend. 5 >>>> Inside a dog it's too dark to read. 5 >>>> Next to a dog a book is man's best friend. 6 >>>> Inside a dog it's too dark to read. 6 >>>> Next to a dog a book is man's best friend. 7 >>>> Inside a dog it's too dark to read. 7 >>>> Next to a dog a book is man's best friend. 8 >>>> Inside a dog it's too dark to read. 8 >>>> Next to a dog a book is man's best friend. 9 >>>> Inside a dog it's too dark to read. 9 >>>> >>>> >>>> >>>> -- >>>> users mailing list >>>> users(a)lists.fedoraproject.org >>>> To unsubscribe or change subscription options: >>>> https://admin.fedoraproject.org/mailman/listinfo/users >>>> Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines >>>> Have a question? Ask away: http://ask.fedoraproject.org >> >>

1 0

Re: parallel bash scripts
by Mark LaPierre 27 Mar '12

27 Mar '12

On 03/27/2012 09:37 PM, Mark LaPierre wrote: > Check out the redirection at the end of each command. 1>&2 redirects the > standard out of your child command to the standard error which then > appears in the parent shell. At the end the last & launches your command > into a background shell and then moves on to launch the next command. > The redirections don't care if the command ever terminates. > > The result is that both commands are launched and the parent shell > terminates leaving the standard error attached to the terminal that the > parent was launched in. > > On 03/27/2012 09:08 PM, bruce wrote: >> marklapier(a)aol.com >> >> hey mark.... >> >> what you have, appears to be pretty close to what i had... except my >> tests never ended... the loops are infinite... >> >> can i do a fpaste and have you take a look at what i have? >> >> -btuce >> >> >> On Tue, Mar 27, 2012 at 9:00 PM, Mark LaPierre<marklapier(a)aol.com> wrote: >>> On 03/27/2012 05:25 PM, bruce wrote: >>>> >>>> hi. >>>> >>>> got a couple of test bash scripts. >>>> >>>> dog.sh, cat.sh >>>> each script runs the underlying php in an endless loop. >>>> >>>> I'm trying to figure out how to run the scripts in parallel, from the >>>> same parent shell script. something like: >>>> >>>> test.sh >>>> >>>> >>>> where dog.sh would be : >>>> -------------------------------- >>>> while true >>>> do >>>> pgrep dog >>>> if [ $? -ne 0 ] >>>> then >>>> /dog.php >>>> fi >>>> sleep 5 >>>> done >>>> >>>> my current tests, run dog.sh, which runs the dog.php ... but the test >>>> never gets to run cat.sh >>>> >>>> thoughts/comments... >>>> >>>> thanks >>> >>> Hey Bruce, >>> >>> Do you mean to run these subprograms in parallel or in series? >>> >>> cat.sh >>> #! /bin/bash >>> >>> CAT=0 >>> until [ $CAT -eq 10 ] >>> do >>> echo "Inside a dog it's too dark to read. $CAT" >>> CAT=$[$CAT + 1] >>> sleep 2 >>> done >>> >>> >>> dog.sh >>> #! /bin/bash >>> >>> DOG=0 >>> until [ $DOG -eq 10 ] >>> do >>> echo "Next to a dog a book is man's best friend. $DOG" >>> DOG=$[$DOG + 1] >>> sleep 2 >>> done >>> >>> >>> test.sh >>> #! /bin/sh >>> >>> /home/mlapier/test/dog.sh 1>&2& >>> /home/mlapier/test/cat.sh 1>&2& >>> >>> >>> [mlapier@mushroom test]$ ./test.sh >>> [mlapier@mushroom test]$ Next to a dog a book is man's best friend. 0 >>> Inside a dog it's too dark to read. 0 >>> Next to a dog a book is man's best friend. 1 >>> Inside a dog it's too dark to read. 1 >>> Next to a dog a book is man's best friend. 2 >>> Inside a dog it's too dark to read. 2 >>> Next to a dog a book is man's best friend. 3 >>> Inside a dog it's too dark to read. 3 >>> Next to a dog a book is man's best friend. 4 >>> Inside a dog it's too dark to read. 4 >>> Next to a dog a book is man's best friend. 5 >>> Inside a dog it's too dark to read. 5 >>> Next to a dog a book is man's best friend. 6 >>> Inside a dog it's too dark to read. 6 >>> Next to a dog a book is man's best friend. 7 >>> Inside a dog it's too dark to read. 7 >>> Next to a dog a book is man's best friend. 8 >>> Inside a dog it's too dark to read. 8 >>> Next to a dog a book is man's best friend. 9 >>> Inside a dog it's too dark to read. 9 >>> >>> >>> >>> -- >>> users mailing list >>> users(a)lists.fedoraproject.org >>> To unsubscribe or change subscription options: >>> https://admin.fedoraproject.org/mailman/listinfo/users >>> Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines >>> Have a question? Ask away: http://ask.fedoraproject.org >

1 0

minimal install CD versus Pentium 4
by Michael Hennebry 27 Mar '12

27 Mar '12

I've tried to install F16 using its minimal CD and its iso file. The CD will not boot. Has anyone gotten this combination to boot? Most recently I used tab and backspaced over quiet. Since I didn't have a running system running, I had to copy the result by hand: [1.429346] ---cut here--- kernel BUG at arch/x86/kernel/traps.c:531! invalid opcode 0000 [#1] SMP Modules linked in: Pid 1: 1, comm: swapper Not tainted 3.1.0-7.f16.i686 #1 Daktech**DT6000* /D865GBF EIP: 0060:[<c0811b4e>] EFLAGS: 00010006 CPU: 1 EIP is at do_nmi+0x22/0x280 EAX: ec498000 EBX: 9ed363a7 ECX: 00000000 EDX: ec499b0c ESI: ec499b38 EDI: c0a5c746 EBP: ec499b30 ESP: ec499b0c DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 Process swapper (pid: 1, t:=0c490000 task=ec49000 task.ti=ec498?00) Stack: 00000000 00000002 00000001 ffff0005 ec499bc8 9ed3c63a7 ?f8ce356 c0a5c746 ecbcb8c0 c08117bc 9ed363a7 00000000 ff8ce356 ?0a5c746 ecbcb8c0 000054b2 c05c00b7 0000007b 000000e0 000054b2 ?05c9745 call trace Every line began [1.429346] The *s represent little circles. The question marks are the result of wrap-around and my screen apparently being 79.3 characters wide. I expect that they should be the same as the hex digits following them. The CD is not bad. It's an exact copy of the iso from which it was made. I read it back and did a diff to make sure. The iso's sha256sum is a4a7cc83163a2cecf1420c64d48123d0887e165aa17a72b8920f8d70bf003a65 . Any ideas? In another thread, it was suggested that this bugzilla entry might be pertinent: https://bugzilla.redhat.com/show_bug.cgi?id=730007 While I can certainly see the relationship, it's not obviously the same thing. Anyone know for sure? -- Michael hennebry(a)web.cs.ndsu.NoDak.edu "On Monday, I'm gonna have to tell my kindergarten class, whom I teach not to run with scissors, that my fiance ran me through with a broadsword." -- Lily

8 12

Processor time consumption
by Howard Howell 27 Mar '12

27 Mar '12

Hi, everyone, My latest boondoggle is a multiple one. I loaded tracker to fool with it a bit, and now it is always on, I looked up the man pages and checked the commands, so I know I can turn off the various processes using tracker-control -t. But how do I keep it from starting? I used to know how to edit some of the startup configuration, but now I find that what I remember is wayyyyyy out of date. Any guidance would be appreciated. With that out of the way, and the tracker processes stopped, I have three remaining problems... 1. I have a spreadsheet I want to use to create charts for a book I am writting, so I run the spreadsheet in libre office. 2. I then open the book in libre office. Now soffice.bin is consuming 99+ percent of the processor time and everything grinds to a stop. 3. I open evolution, retrieve my mail, everything is normal. I begin to edit a messge adn about 3 lines in, it grinds to a halt, soffice is not running. After about 30 seconds it resumes working. Since I would suppose that most of you use Libre office, and this is not likely happening as I see no comments to that effect, I suspect that the problem is mine alone. This makes me think configuration or other error. But I already have lots of errors from various processes, but none from soffice that I can tell. These errors do not appear to effect about 30 other applications I run, and while I know that this doesn't mean they won't affect soffice, I think that is unlikely, so my guess is that there is some issue in the soffice configuration, or that I am missing some file or link. Any thoughts? The evolution problem could be related to some of the errors, dealing with dbus, and I am working on tracking those down. Any pointers appreciated. (I arrived at this due to timestamp relationships.) Regards, Les H

2 1

FC15/16 NetworkManager drops network connection
by Symersky, Henry 27 Mar '12

27 Mar '12

Dear all, I may have found answer to my previous question regarding disabled login and the message: "could not update ICEauthority file /home/username/.ICEauthority" I found that whenever I logout as root, the NetworkManager closes ethernet connection. So whenever I want to login as normal user, it cannot access the home directory, because the network is down. Can someone please explain to me how to configure the NetworkManager in order to KEEP ETEHERNET CONNECTTION EVEN AFTER LOGGING OUT ?? Thanks. Henry

2 2

Fedora and the horrible EDID checksum issue with various video cards
by Mark Haney 27 Mar '12

27 Mar '12

Hi guys, got a question I hope someone can answer pretty quickly. I've been using kubuntu on various slower systems the last couple of years, but since Canonical is dropping KDE I want to shift them all back to Fedora. I've noticed that F16 flies on hardware that F12 just killed so I'm willing to give it a shot. My question is this: I've had an issue with newer ubuntu kernels (3.0.14,.15 and .16 specifically) that just implode on my video card with > [drm:drm_edid_block_valid] *ERROR* EDID checksum is invalid, remainder is 128 repeatedly, pegging my proc and rendering the system unusable. Has the 3.1+ kernels addressed this issue? Is it still hanging out there? What exactly is the status of this in the Fedora realm? Right now, I have a system I can use (with the 0.13 kernel) but I don't want to upgrade if it leaves me with an unusable system. -- Mark Haney Software Developer/Consultant AB Emblem markh(a)abemblem.com Linux marius.homelinux 3.3.0-4.fc16.x86_64 GNU/Linux

3 4

Unable to mount (samba) location on Fedora 17
by Mateusz Marzantowicz 27 Mar '12

27 Mar '12

Hello, I have recently installed Fedora 17, one on my EeePC and second inside KVM virtual machine on my desktop. The problem is that I cannot access my shared folder, exported on my Ubuntu Samba Server from Nautilus on Fedora. When I use smbclient command - it works! I am also able to mount my samba share from command line on both EeePC and KVM. I have all recent updates installed on all systems. I have disabled firewall and SELinux but the problem persists and it seems that it is related to Nautilus and not anything else. I have different passwords set on my samba server and clients (same user name), but Nautilus does not ask me for neither user name nor password. Is there anything I can do/set/etc. to resolve this issue? Thanks for any reply, Mateusz Marzantowicz

1 0

fc16-.ICEauthority
by Symersky, Henry 26 Mar '12

26 Mar '12

Dear all, A system with fc16 does not allow normal users to login: "Could not update ICEauthority file /home/username/.ICEauthority". The /home/user is NFS-connected, and it works fine for other systems. Any suggestions how to fix it ? Thanks. Henry

3 4

still having update issues with gcc
by Howard Howell 26 Mar '12

26 Mar '12

HI, everyone, I am still having the update issue with gcc. I will follow this message with the error received from "yum update". Also my processor is periodically locking up. Is anyone else experiencing either of these errors? When it locks up, it is running one of the cores at or near 100%, and the dbus daemon is running at 54%, and if I click the x box on system monitor, the window remains open, but the process appears to have died. I am trying to figure out if these are related or separate issues that I have to resolve. Can anyone offer any guidance, and as to the update of gcc, there was an earlier email where someone indicated that a update was needed but not available yet. Any help on that front either? Regards, Les H # yum install gcc-c++.x86_64 Loaded plugins: langpacks, presto, refresh-packagekit Setting up Install Process Resolving Dependencies --> Running transaction check ---> Package gcc-c++.x86_64 0:4.6.2-1.fc16 will be updated --> Processing Dependency: gcc-c++ = 4.6.2 for package: clang-2.9-6.fc16.x86_64 ---> Package gcc-c++.x86_64 0:4.6.3-2.fc16 will be an update --> Processing Dependency: libstdc++-devel = 4.6.3-2.fc16 for package: gcc-c++-4.6.3-2.fc16.x86_64 --> Processing Dependency: libstdc++ = 4.6.3-2.fc16 for package: gcc-c ++-4.6.3-2.fc16.x86_64 --> Processing Dependency: gcc = 4.6.3-2.fc16 for package: gcc-c ++-4.6.3-2.fc16.x86_64 --> Running transaction check ---> Package gcc.x86_64 0:4.6.2-1.fc16 will be updated --> Processing Dependency: gcc = 4.6.2-1.fc16 for package: gcc-java-4.6.2-1.fc16.x86_64 --> Processing Dependency: gcc = 4.6.2 for package: libtool-2.4-8.fc16.x86_64 ---> Package gcc.x86_64 0:4.6.3-2.fc16 will be an update --> Processing Dependency: libgomp = 4.6.3-2.fc16 for package: gcc-4.6.3-2.fc16.x86_64 --> Processing Dependency: cpp = 4.6.3-2.fc16 for package: gcc-4.6.3-2.fc16.x86_64 --> Processing Dependency: libgcc >= 4.6.3-2.fc16 for package: gcc-4.6.3-2.fc16.x86_64 ---> Package gcc-c++.x86_64 0:4.6.2-1.fc16 will be updated --> Processing Dependency: gcc-c++ = 4.6.2 for package: clang-2.9-6.fc16.x86_64 ---> Package libstdc++.i686 0:4.6.2-1.fc16 will be updated ---> Package libstdc++.x86_64 0:4.6.2-1.fc16 will be updated ---> Package libstdc++.i686 0:4.6.3-2.fc16 will be an update ---> Package libstdc++.x86_64 0:4.6.3-2.fc16 will be an update ---> Package libstdc++-devel.x86_64 0:4.6.2-1.fc16 will be updated ---> Package libstdc++-devel.x86_64 0:4.6.3-2.fc16 will be an update --> Running transaction check ---> Package cpp.x86_64 0:4.6.2-1.fc16 will be updated ---> Package cpp.x86_64 0:4.6.3-2.fc16 will be an update ---> Package gcc-c++.x86_64 0:4.6.2-1.fc16 will be updated --> Processing Dependency: gcc-c++ = 4.6.2 for package: clang-2.9-6.fc16.x86_64 ---> Package gcc-java.x86_64 0:4.6.2-1.fc16 will be updated ---> Package gcc-java.x86_64 0:4.6.3-2.fc16 will be an update --> Processing Dependency: libgcj-devel = 4.6.3-2.fc16 for package: gcc-java-4.6.3-2.fc16.x86_64 --> Processing Dependency: libgcj = 4.6.3-2.fc16 for package: gcc-java-4.6.3-2.fc16.x86_64 ---> Package libgcc.i686 0:4.6.2-1.fc16 will be updated ---> Package libgcc.x86_64 0:4.6.2-1.fc16 will be updated ---> Package libgcc.i686 0:4.6.3-2.fc16 will be an update ---> Package libgcc.x86_64 0:4.6.3-2.fc16 will be an update ---> Package libgomp.x86_64 0:4.6.2-1.fc16 will be updated ---> Package libgomp.x86_64 0:4.6.3-2.fc16 will be an update ---> Package libtool.x86_64 0:2.4-8.fc16 will be updated ---> Package libtool.x86_64 0:2.4-9.fc16 will be an update --> Running transaction check ---> Package gcc-c++.x86_64 0:4.6.2-1.fc16 will be updated --> Processing Dependency: gcc-c++ = 4.6.2 for package: clang-2.9-6.fc16.x86_64 ---> Package libgcj.x86_64 0:4.6.2-1.fc16 will be updated ---> Package libgcj.x86_64 0:4.6.3-2.fc16 will be an update ---> Package libgcj-devel.x86_64 0:4.6.2-1.fc16 will be updated ---> Package libgcj-devel.x86_64 0:4.6.3-2.fc16 will be an update --> Finished Dependency Resolution Error: Package: clang-2.9-6.fc16.x86_64 (@updates) Requires: gcc-c++ = 4.6.2 Removing: gcc-c++-4.6.2-1.fc16.x86_64 (@koji-override-0/$releasever) gcc-c++ = 4.6.2-1.fc16 Updated By: gcc-c++-4.6.3-2.fc16.x86_64 (updates) gcc-c++ = 4.6.3-2.fc16 You could try using --skip-broken to work around the problem You could try running: rpm -Va --nofiles --nodigest

3 3

NetworkManager serious usablity challenges
by Robert Moskowitz 26 Mar '12

26 Mar '12

F16, Gnome3. (sent also to gnome NM list) I am at a friend's house in Amsterdam trying to get connect to his wireless and it is failing, so this message SHOULD go out when I get to the KLM lounge tommorrow (that was working friday)... A major defiency is the loss of deleting SSID configurations. There is no 'delete' feature anymore in the Network Settings panel. There is a network near here that has the same SSID as at another friends (let's call it NETGEAR), but this one has a different password that I do not know. Doesn't matter, NM keeps trying to connect and asks me for a different password when it fails. I have no way (or found no way), to delete or even deactivate this SSID from NM. So it keeps trying and trying. Connection to my friend's wireless SEEMs to be a DHCP problem. I have this hunch by watching /var/log/messages; this 'new' network manager does not tell me why it is failing. Now I shoud preface the next part with I work in 802.11 standards. Right now I am active in 802.11ai (FIA), so I KNOW the .11 state machine. Is the problem in initial connection (AUTH,ASSOC); note it is possible to be receiving BEACONs, but be too far to actually ASSOCIATE with an AP. The user should be told the problem is here. Or is it a bad password; well that is the guess when it presents the dialog for the password but i KNOW the password is correct. Oh, perhaps the problem is DHCPv4 (or v6?) and since there is no way to tell the user to fix the DHCP allocation in the router, the poor user gets asked to try a different password? You KNOW what the failure is. PLEASE give some information as to which step things stop at. Plus change the icon from that strange ... thing to something showing trying and trying what? (ASSOC, SECURE, ADDRESS). Also be so informed that when 11ai gets done (you do have 2 years) we are going to do all this in a couple/few roundtrips. My proposal does the whole shabang in 2. The AUTH starts the securing and the ASSOC finishes the securing and does the addressing, though there are times where addressing extends the ASSOC for another roundtrip. Now back to feature loss over Gnome2 NM.... I cannot turn off wireless from the NM pulldown if it is currently trying (and really failing) to connect. I have to open the Network Settings dialog and turn off wireless there. While attempting to connect the on/off switch is replaced with the text 'connecting'. I know that, I want to stop it trying to connect, and the only way to do that is turn the wireless off. Does turning NM off turn off the wireless radio? I have a Lenovo x120 and it does not have a wireless radio switch and on airplanes, I like to turn off my radio and save battery. All I can do is turn off NM, but I have no way of knowing WHAT is being turned off! Enough challenges for tonight. Tomorrow it is off to the airport and hopefully connectivity there on my way to the IETF meeting. (yes I am a standards guy).

6 12

← Newer
1
...
6
7
8
9
10
11
12
...
31
Older →

Jump to page: