Re: Re: Marvell 88E8071 PCI-E Gigabit Ethernet / packet_write_wait: Connection to ...: Broken pipe
Thank you for the prompt reply! Unfortunatelly, I can't open a local console, the box has VGA output and I have no VGA monitor. So, SSH is the only option. But I do can boot a clean SQUEEZE system, mount the root file system of the broken JESSIE system and download log files. What information should I gather? -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/5571ef82.7080...@gmx.de
Marvell 88E8071 PCI-E Gigabit Ethernet / packet_write_wait: Connection to ...: Broken pipe
Hi, I've got an Acer Cube and experience the following problem: Multiple Debian systems are installed: lenny, squeeze, wheezy, jessie; all four are almost bare base systems. The first two, lenny and squeeze, funktion without any problems. The last two, wheezy and jessie, start without problems, and also let me to connect over SSH. But after 1-2 minutes the client terminal freezes and the outputs this error message: packet_write_wait: Connection to 192.168.1.2: Broken pipe. The network activity LED on the box is flushing permanently. After that, no more connections are accepted, the only way to access the system is a reboot using the power button. This command, if executed right after login, seems to delay the problem by a minute or so: # ethtool -K eth0 tso off Any help would be greatly appreciated! --- Information obtained while running SQUEEZE: root@vega:~# lspci 00:00.0 Host bridge: Intel Corporation N10 Family DMI Bridge 00:02.0 VGA compatible controller: Intel Corporation N10 Family Integrated Graphics Controller 00:1a.0 USB Controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #4 (rev 02) 00:1a.1 USB Controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #5 (rev 02) 00:1a.2 USB Controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #6 (rev 02) 00:1a.7 USB Controller: Intel Corporation 82801I (ICH9 Family) USB2 EHCI Controller #2 (rev 02) 00:1c.0 PCI bridge: Intel Corporation 82801I (ICH9 Family) PCI Express Port 1 (rev 02) 00:1c.5 PCI bridge: Intel Corporation 82801I (ICH9 Family) PCI Express Port 6 (rev 02) 00:1d.0 USB Controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #1 (rev 02) 00:1d.1 USB Controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #2 (rev 02) 00:1d.2 USB Controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #3 (rev 02) 00:1d.7 USB Controller: Intel Corporation 82801I (ICH9 Family) USB2 EHCI Controller #1 (rev 02) 00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev 92) 00:1f.0 ISA bridge: Intel Corporation 82801IR (ICH9R) LPC Interface Controller (rev 02) 00:1f.2 SATA controller: Intel Corporation 82801IR/IO/IH (ICH9R/DO/DH) 6 port SATA AHCI Controller (rev 02) 00:1f.3 SMBus: Intel Corporation 82801I (ICH9 Family) SMBus Controller (rev 02) 02:00.0 Ethernet controller: Marvell Technology Group Ltd. 88E8071 PCI-E Gigabit Ethernet Controller (rev 16) root@vega:~# ethtool eth0 Settings for eth0: Supported ports: [ TP ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Advertised pause frame use: No Advertised auto-negotiation: Yes Speed: 100Mb/s Duplex: Full Port: Twisted Pair PHYAD: 0 Transceiver: internal Auto-negotiation: on MDI-X: Unknown Supports Wake-on: pg Wake-on: d Current message level: 0x00ff (255) Link detected: yes root@vega:~# ethtool -i eth0 driver: sky2 version: 1.28 firmware-version: N/A bus-info: :02:00.0 root@vega:~# ethtool -i eth0 driver: sky2 version: 1.28 firmware-version: N/A bus-info: :02:00.0 root@vega:~# ethtool -k eth0 Offload parameters for eth0: rx-checksumming: on tx-checksumming: on scatter-gather: on tcp-segmentation-offload: on udp-fragmentation-offload: off generic-segmentation-offload: on generic-receive-offload: off large-receive-offload: off ntuple-filters: off receive-hashing: off --- Information obtained while running JESSIE: root@vega:~# lspci 00:00.0 Host bridge: Intel Corporation Atom Processor D4xx/D5xx/N4xx/N5xx DMI Bridge 00:02.0 VGA compatible controller: Intel Corporation Atom Processor D4xx/D5xx/N4xx/N5xx Integrated Graphics Controller 00:1a.0 USB controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #4 (rev 02) 00:1a.1 USB controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #5 (rev 02) 00:1a.2 USB controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #6 (rev 02) 00:1a.7 USB controller: Intel Corporation 82801I (ICH9 Family) USB2 EHCI Controller #2 (rev 02) 00:1c.0 PCI bridge: Intel Corporation 82801I (ICH9 Family) PCI Express Port 1 (rev 02) 00:1c.5 PCI bridge: Intel Corporation 82801I (ICH9 Family) PCI Express Port 6 (rev 02) 00:1d.0 USB controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #1 (rev 02) 00:1d.1 USB controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #2 (rev 02) 00:1d.2 USB controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #3 (rev 02) 00:1d.7 USB
Re: Marvell 88E8071 PCI-E Gigabit Ethernet / packet_write_wait: Connection to ...: Broken pipe
On Fri, 2015-06-05 at 18:39 +0200, SkyFish wrote: Hi, I've got an Acer Cube and experience the following problem: Multiple Debian systems are installed: lenny, squeeze, wheezy, jessie; all four are almost bare base systems. The first two, lenny and squeeze, funktion without any problems. The last two, wheezy and jessie, start without problems, and also let me to connect over SSH. But after 1-2 minutes the client terminal freezes and the outputs this error message: packet_write_wait: Connection to 192.168.1.2: Broken pipe. The network activity LED on the box is flushing permanently. After that, no more connections are accepted, the only way to access the system is a reboot using the power button. The best way to gather information would probably be to login (locally or with another network card) and see what happens during the failure. Messages in dmesg, kernel oops etc. Since you have a working setup it should be possible to bisect and find where the bug was introduced (if it is a kernel problem). -- Cheers, Sven Arvidsson http://www.whiz.se PGP Key ID 6FAB5CD5 signature.asc Description: This is a digitally signed message part
Re: Re: Marvell 88E8071 PCI-E Gigabit Ethernet / packet_write_wait: Connection to ...: Broken pipe
On Fri, 2015-06-05 at 20:50 +0200, SkyFish wrote: Thank you for the prompt reply! Unfortunatelly, I can't open a local console, the box has VGA output and I have no VGA monitor. Buy an adapter! ;) So, SSH is the only option. But I do can boot a clean SQUEEZE system, mount the root file system of the broken JESSIE system and download log files. What information should I gather? I think /var/log/syslog should have everything. -- Cheers, Sven Arvidsson http://www.whiz.se PGP Key ID 6FAB5CD5 signature.asc Description: This is a digitally signed message part
dpkg-deb: error: subprocess paste was killed by signal (Broken pipe)
Hi people! I really don't know, how to solve this problem. Does any of you have an idea ?! root@babe3306:/home/tamer# apt-get install -f Reading package lists... Done Building dependency tree Reading state information... Done Correcting dependencies... Done The following extra packages will be installed: libsope1 The following NEW packages will be installed: libsope1 0 upgraded, 1 newly installed, 0 to remove and 54 not upgraded. 5 not fully installed or removed. Need to get 0 B/1883 kB of archives. After this operation, 9527 kB of additional disk space will be used. Do you want to continue? [Y/n] Y (Reading database ... 220104 files and directories currently installed.) Preparing to unpack .../libsope1_2.2.5-1_amd64.deb ... Unpacking libsope1 (2.2.5-1) ... dpkg: error processing archive /var/cache/apt/archives/libsope1_2.2.5-1_amd64.deb (--unpack): trying to overwrite '/usr/lib/libDOM.so.4.9.24', which is also in package libsope-xml4.9 4.9.r1664.20140307 dpkg-deb: error: subprocess paste was killed by signal (Broken pipe) Errors were encountered while processing: /var/cache/apt/archives/libsope1_2.2.5-1_amd64.deb E: Sub-process /usr/bin/dpkg returned an error code (1) root@babe3306:/home/tamer# I have latest debian testing running, and don't know how to get that one solved. Thanks! -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/53b20a26.3050...@googlemail.com
Re: dpkg-deb: error: subprocess paste was killed by signal (Broken pipe)
On Tue, 01 Jul 2014, Tamer Higazi wrote: root@babe3306:/home/tamer# apt-get install -f [...] Preparing to unpack .../libsope1_2.2.5-1_amd64.deb ... Unpacking libsope1 (2.2.5-1) ... dpkg: error processing archive /var/cache/apt/archives/libsope1_2.2.5-1_amd64.deb (--unpack): trying to overwrite '/usr/lib/libDOM.so.4.9.24', which is also in package libsope-xml4.9 4.9.r1664.20140307 dpkg-deb: error: subprocess paste was killed by signal (Broken pipe) Errors were encountered while processing: /var/cache/apt/archives/libsope1_2.2.5-1_amd64.deb E: Sub-process /usr/bin/dpkg returned an error code (1) root@babe3306:/home/tamer# libsope-xml4.9 was never in Debian. I'm going to guess that you got it from Ubuntu or somewhere else. Remove it, and then you can install libsope1. Alternatively, use apt-get -o DPkg::Options::=--force-overwrite install libsope1 to enable libDOM.so.4.9.24 to be overwritten. -- Don Armstrong http://www.donarmstrong.com Where am I? THE VILLAGE. What do you want? INFORMATION. Which side are you on? THAT WOULD BE TELLING. WE WANT INFORMATION. INFORMATION. INFORMATION. You won't get it! BY HOOK OR BY CROOK, WE WILL. Who are you? THE NEW NUMBER 2. Who is Number 1? YOU ARE NUMBER 6. I am not a number! I am a free man! HAHAHAHAHAHA. -- Patrick McGoohan as Number 6 with Number 2 in The Prisoner -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140701015748.gr9...@teltox.donarmstrong.com
Re: dpkg-deb: error: subprocess paste was killed by signal (Broken pipe)
On Tue, Jul 01, 2014 at 03:08:54AM +0200, Tamer Higazi wrote: Hi people! I really don't know, how to solve this problem. Does any of you have an idea ?! root@babe3306:/home/tamer# apt-get install -f Preparing to unpack .../libsope1_2.2.5-1_amd64.deb ... Unpacking libsope1 (2.2.5-1) ... dpkg: error processing archive /var/cache/apt/archives/libsope1_2.2.5-1_amd64.deb (--unpack): trying to overwrite '/usr/lib/libDOM.so.4.9.24', which is also in package libsope-xml4.9 4.9.r1664.20140307 There's your problem, obviously two packages can't provide the same file. Where did you get libsope-xml4.9 from? https://packages.debian.org/search?keywords=libsope-xml4.9searchon=namessuite=allsection=all produces: Sorry, your search gave no results -- If you're not careful, the newspapers will have you hating the people who are being oppressed, and loving the people who are doing the oppressing. --- Malcolm X -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140701020001.GA7901@tal
Re: Inexplicable broken pipe entre dos subredes bajo mismo LAN
El 24 de marzo de 2014, 23:54, Santiago Liz lizsa...@gmail.com escribió: El día 23 de marzo de 2014, 6:32, Roberto i32lelor.deb...@gmail.com escribió: Administro una red LAN donde siempre que puedo aplico Debian a la ecuación. El servidor que hace de gateway hace de firewall y registra log determinado trafico. Tiene dos interfaces br0 y br0:1 con 192.168.0.2 y 192.168.2.1. Uso un alias por no disponer de dos tarjetas de red, y uso br0 porque en dicho equipo uso VirtualBox. Tengo un servidor con jenkins(Debian) en la 192.168.0.8 el cual conecta por ssh a dos nodos esclavos, 192.168.2.19(Centos) y 192.168.2.21(Ubuntu). El primero tiene como gateway a 192.168.0.2 y el segundo a 192.168.2.1. Con Ubuntu ningún problema con el trafico de red pero con Centos aleatoriamente ocurre un broken pipe de una sesión ssh, lo único que se me ocurrió despues de días dando cabezasos es pasar ese nodo Centos a la subred con ip 192.168.0.9 y todo arreglado. Abajo pegaré el log de un broken pipe. ¿Hay alguna idea para en esta situación se produzca un broken pipe? Y he configurado Centos para no usar Gssapi, DenyDNS, y lo típico en cliente para evitar un time out. Si las dos redes que tienes definidas comparten físicamente el mismos medio, es probable que el cliente ssh en una de las redes al conectarse al servidor en la otra red, envíe los paquetes hacia la mac address de su default gateway y tu firewall/router reenvíe esos paquetes al servidor ssh. Lo lógico sería pensar que el servidor ssh al establecer la conexión envíe los paquetes corresponedientes de regreso hacia el cliente también por medio de su default gateway. Por alguna razón (que no me puse a investigar), al compartir ambos (cliente y servidor) el mismo medio físico, uno de los dos aprende la mac address del otro y termina enviando los paquetes en forma directa sin intervención del defualt gateway, esto ocasiona un camino asimétrico, que ante la presencia de un firewall statefull provoca el corte de comunicación. jenkins@192.168.2.19's password: debug3: packet_send2: adding 64 (len 60 padlen 4 extra_pad 64) debug2: we sent password packet, wait for reply debug1: Authentication succeeded (password). Authenticated to 192.168.2.19 ([192.168.2.19]:22). debug1: channel 0: new [client-session] debug3: ssh_session2_open: channel_new: 0 debug2: channel 0: send open debug1: Requesting no-more-sessi...@openssh.com debug1: Entering interactive session. debug2: callback start debug2: client_session2_setup: id 0 debug2: fd 3 setting TCP_NODELAY debug3: packet_set_tos: set IP_TOS 0x10 debug2: channel 0: request pty-req confirm 1 debug1: Sending environment. debug3: Ignored env SHELL debug3: Ignored env TERM debug3: Ignored env USER debug3: Ignored env LS_COLORS debug3: Ignored env MAIL debug3: Ignored env PATH debug3: Ignored env PWD debug1: Sending env LANG = es_ES.UTF-8 debug2: channel 0: request env confirm 0 debug3: Ignored env SHLVL debug3: Ignored env HOME debug3: Ignored env LOGNAME debug3: Ignored env _ debug2: channel 0: request shell confirm 1 debug2: callback done debug2: channel 0: open confirm rwindow 0 rmax 32768 Write failed: Broken pipe -- To UNSUBSCRIBE, email to debian-user-spanish-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/532eaa40.5060...@gmail.com Saludos, Santiago.- Pues efectivamente, las rutas son asimétrica como he mostrado en los resultados en email anterior. Y he encontrado como las distros RHEL/Centos no aceptan el tráfico asimétrico a partir de las release 5. En esta url hablan de ello https://access.redhat.com/site/solutions/53031 . Esto lo desconocía ya que suelo montar todo lo que puedo bajo Debian Ya solo me queda investigar el porqué en la misma LAN se produce éste tráfico asimétrico. -- To UNSUBSCRIBE, email to debian-user-spanish-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/calzppqo-aqj7jdp0px67y824abn7eg340-_hf+vtq-+k0ne...@mail.gmail.com
Re: Inexplicable broken pipe entre dos subredes bajo mismo LAN
El 25/03/2014 16:28, Santiago Liz escribió: El día 25 de marzo de 2014, 5:42, Roberto i32lelor.deb...@gmail.com escribió: El 24/03/2014 23:54, Santiago Liz escribió: El día 23 de marzo de 2014, 6:32, Roberto i32lelor.deb...@gmail.com escribió: Administro una red LAN donde siempre que puedo aplico Debian a la ecuación. El servidor que hace de gateway hace de firewall y registra log determinado trafico. Tiene dos interfaces br0 y br0:1 con 192.168.0.2 y 192.168.2.1. Uso un alias por no disponer de dos tarjetas de red, y uso br0 porque en dicho equipo uso VirtualBox. Tengo un servidor con jenkins(Debian) en la 192.168.0.8 el cual conecta por ssh a dos nodos esclavos, 192.168.2.19(Centos) y 192.168.2.21(Ubuntu). El primero tiene como gateway a 192.168.0.2 y el segundo a 192.168.2.1. Con Ubuntu ningún problema con el trafico de red pero con Centos aleatoriamente ocurre un broken pipe de una sesión ssh, lo único que se me ocurrió despues de días dando cabezasos es pasar ese nodo Centos a la subred con ip 192.168.0.9 y todo arreglado. Abajo pegaré el log de un broken pipe. ¿Hay alguna idea para en esta situación se produzca un broken pipe? Y he configurado Centos para no usar Gssapi, DenyDNS, y lo típico en cliente para evitar un time out. Si las dos redes que tienes definidas comparten físicamente el mismos medio, es probable que el cliente ssh en una de las redes al conectarse al servidor en la otra red, envíe los paquetes hacia la mac address de su default gateway y tu firewall/router reenvíe esos paquetes al servidor ssh. Lo lógico sería pensar que el servidor ssh al establecer la conexión envíe los paquetes corresponedientes de regreso hacia el cliente también por medio de su default gateway. Por alguna razón (que no me puse a investigar), al compartir ambos (cliente y servidor) el mismo medio físico, uno de los dos aprende la mac address del otro y termina enviando los paquetes en forma directa sin intervención del defualt gateway, esto ocasiona un camino asimétrico, que ante la presencia de un firewall statefull provoca el corte de comunicación. jenkins@192.168.2.19's password: debug3: packet_send2: adding 64 (len 60 padlen 4 extra_pad 64) debug2: we sent password packet, wait for reply debug1: Authentication succeeded (password). Authenticated to 192.168.2.19 ([192.168.2.19]:22). debug1: channel 0: new [client-session] debug3: ssh_session2_open: channel_new: 0 debug2: channel 0: send open debug1: Requesting no-more-sessi...@openssh.com debug1: Entering interactive session. debug2: callback start debug2: client_session2_setup: id 0 debug2: fd 3 setting TCP_NODELAY debug3: packet_set_tos: set IP_TOS 0x10 debug2: channel 0: request pty-req confirm 1 debug1: Sending environment. debug3: Ignored env SHELL debug3: Ignored env TERM debug3: Ignored env USER debug3: Ignored env LS_COLORS debug3: Ignored env MAIL debug3: Ignored env PATH debug3: Ignored env PWD debug1: Sending env LANG = es_ES.UTF-8 debug2: channel 0: request env confirm 0 debug3: Ignored env SHLVL debug3: Ignored env HOME debug3: Ignored env LOGNAME debug3: Ignored env _ debug2: channel 0: request shell confirm 1 debug2: callback done debug2: channel 0: open confirm rwindow 0 rmax 32768 Write failed: Broken pipe -- To UNSUBSCRIBE, email to debian-user-spanish-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/532eaa40.5060...@gmail.com Saludos, Santiago.- Es decir, que aun teniendo bien la configuración de cada equipo apuntando al gateway 192.168.0.2 en el caso de LAN puede realizar el camino directo esto lo voy a ver con mtrr y os detallo que ocurre. Informar, que hay otro gateway 192.168.0.1, pero este no es usado por ninguno de estos nodos y sólo por el servidor central para el tráfico hacia internet. Es que en principio no es correcto que un equipo de la red 192.168.2.0/24 tenga como gateway una IP de la red 192.168.0.0/24. En cuanto a las pruebas a hacer, el mtr no te va a ayudar mucho, lo mejor un tcpdump al momento de hacer las pruebas de conexión y algún listado de arp en los equipos involucrados. -- To UNSUBSCRIBE, email to debian-user-spanish-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/5331418d.6010...@gmail.com Saludos, Santiago.- Claro que no tengo configurado el gateway de cada nodo en una subred diferente. Interesante el resultado del tracepath, en un sentido pasa por el gateway y a la inversa la conexión es directa sin pasar por su gateway que seria la 192.168.0.2. Nodo 192.168.2.19: tracepath 192.168.0.7 1: 192.168.2.19 0.287ms pmtu 1500 1: 192.168.2.1 5.355ms 1: 192.168.2.1 2.248ms 2: 192.168.0.7 6.605ms reached Resume
Re: Inexplicable broken pipe entre dos subredes bajo mismo LAN
El 24/03/2014 11:34, Francesc Guitart escribió: El 23/03/2014 10:32, Roberto escribió: Administro una red LAN donde siempre que puedo aplico Debian a la ecuación. El servidor que hace de gateway hace de firewall y registra log determinado trafico. Tiene dos interfaces br0 y br0:1 con 192.168.0.2 y 192.168.2.1. Uso un alias por no disponer de dos tarjetas de red, y uso br0 porque en dicho equipo uso VirtualBox. Tengo un servidor con jenkins(Debian) en la 192.168.0.8 el cual conecta por ssh a dos nodos esclavos, 192.168.2.19(Centos) y 192.168.2.21(Ubuntu). El primero tiene como gateway a 192.168.0.2 y el segundo a 192.168.2.1. Dices que el CentOS tiene IP 192.168.2.19 y gateway 192.168.0.2. ¿La gateway del CentOS no debería ser también 192.168.2.1 si está en dicha red? Con Ubuntu ningún problema con el trafico de red pero con Centos aleatoriamente ocurre un broken pipe de una sesión ssh, lo único que se me ocurrió despues de días dando cabezasos es pasar ese nodo Centos a la subred con ip 192.168.0.9 y todo arreglado. Abajo pegaré el log de un broken pipe. Como dice Camaleón me parece raro que sea por culpa de una ruta pero creo que la gateway del CentOS no es correcta. Yo la cambiaría a ver si el error se reproduce. ¿Hay alguna idea para en esta situación se produzca un broken pipe? Y he configurado Centos para no usar Gssapi, DenyDNS, y lo típico en cliente para evitar un time out. jenkins@192.168.2.19's password: debug3: packet_send2: adding 64 (len 60 padlen 4 extra_pad 64) debug2: we sent a password packet, wait for reply debug1: Authentication succeeded (password). Authenticated to 192.168.2.19 ([192.168.2.19]:22). debug1: channel 0: new [client-session] debug3: ssh_session2_open: channel_new: 0 debug2: channel 0: send open debug1: Requesting no-more-sessi...@openssh.com debug1: Entering interactive session. debug2: callback start debug2: client_session2_setup: id 0 debug2: fd 3 setting TCP_NODELAY debug3: packet_set_tos: set IP_TOS 0x10 debug2: channel 0: request pty-req confirm 1 debug1: Sending environment. debug3: Ignored env SHELL debug3: Ignored env TERM debug3: Ignored env USER debug3: Ignored env LS_COLORS debug3: Ignored env MAIL debug3: Ignored env PATH debug3: Ignored env PWD debug1: Sending env LANG = es_ES.UTF-8 debug2: channel 0: request env confirm 0 debug3: Ignored env SHLVL debug3: Ignored env HOME debug3: Ignored env LOGNAME debug3: Ignored env _ debug2: channel 0: request shell confirm 1 debug2: callback done debug2: channel 0: open confirm rwindow 0 rmax 32768 Write failed: Broken pipe El gateway del Centos es 192.168.2.1, al igual que el nodo de Ubuntu. Pero la maquina Debian que los virtualiza lleva un bridged br0 con ip en la subred 192.168.0.7 ,pero vamos un bridged deja pasar todo el trafico de sus maquinas virtualizadas. -- To UNSUBSCRIBE, email to debian-user-spanish-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/53313fd5.7050...@gmail.com
Re: Inexplicable broken pipe entre dos subredes bajo mismo LAN
El 24/03/2014 23:54, Santiago Liz escribió: El día 23 de marzo de 2014, 6:32, Roberto i32lelor.deb...@gmail.com escribió: Administro una red LAN donde siempre que puedo aplico Debian a la ecuación. El servidor que hace de gateway hace de firewall y registra log determinado trafico. Tiene dos interfaces br0 y br0:1 con 192.168.0.2 y 192.168.2.1. Uso un alias por no disponer de dos tarjetas de red, y uso br0 porque en dicho equipo uso VirtualBox. Tengo un servidor con jenkins(Debian) en la 192.168.0.8 el cual conecta por ssh a dos nodos esclavos, 192.168.2.19(Centos) y 192.168.2.21(Ubuntu). El primero tiene como gateway a 192.168.0.2 y el segundo a 192.168.2.1. Con Ubuntu ningún problema con el trafico de red pero con Centos aleatoriamente ocurre un broken pipe de una sesión ssh, lo único que se me ocurrió despues de días dando cabezasos es pasar ese nodo Centos a la subred con ip 192.168.0.9 y todo arreglado. Abajo pegaré el log de un broken pipe. ¿Hay alguna idea para en esta situación se produzca un broken pipe? Y he configurado Centos para no usar Gssapi, DenyDNS, y lo típico en cliente para evitar un time out. Si las dos redes que tienes definidas comparten físicamente el mismos medio, es probable que el cliente ssh en una de las redes al conectarse al servidor en la otra red, envíe los paquetes hacia la mac address de su default gateway y tu firewall/router reenvíe esos paquetes al servidor ssh. Lo lógico sería pensar que el servidor ssh al establecer la conexión envíe los paquetes corresponedientes de regreso hacia el cliente también por medio de su default gateway. Por alguna razón (que no me puse a investigar), al compartir ambos (cliente y servidor) el mismo medio físico, uno de los dos aprende la mac address del otro y termina enviando los paquetes en forma directa sin intervención del defualt gateway, esto ocasiona un camino asimétrico, que ante la presencia de un firewall statefull provoca el corte de comunicación. jenkins@192.168.2.19's password: debug3: packet_send2: adding 64 (len 60 padlen 4 extra_pad 64) debug2: we sent password packet, wait for reply debug1: Authentication succeeded (password). Authenticated to 192.168.2.19 ([192.168.2.19]:22). debug1: channel 0: new [client-session] debug3: ssh_session2_open: channel_new: 0 debug2: channel 0: send open debug1: Requesting no-more-sessi...@openssh.com debug1: Entering interactive session. debug2: callback start debug2: client_session2_setup: id 0 debug2: fd 3 setting TCP_NODELAY debug3: packet_set_tos: set IP_TOS 0x10 debug2: channel 0: request pty-req confirm 1 debug1: Sending environment. debug3: Ignored env SHELL debug3: Ignored env TERM debug3: Ignored env USER debug3: Ignored env LS_COLORS debug3: Ignored env MAIL debug3: Ignored env PATH debug3: Ignored env PWD debug1: Sending env LANG = es_ES.UTF-8 debug2: channel 0: request env confirm 0 debug3: Ignored env SHLVL debug3: Ignored env HOME debug3: Ignored env LOGNAME debug3: Ignored env _ debug2: channel 0: request shell confirm 1 debug2: callback done debug2: channel 0: open confirm rwindow 0 rmax 32768 Write failed: Broken pipe -- To UNSUBSCRIBE, email to debian-user-spanish-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/532eaa40.5060...@gmail.com Saludos, Santiago.- Es decir, que aun teniendo bien la configuración de cada equipo apuntando al gateway 192.168.0.2 en el caso de LAN puede realizar el camino directo esto lo voy a ver con mtrr y os detallo que ocurre. Informar, que hay otro gateway 192.168.0.1, pero este no es usado por ninguno de estos nodos y sólo por el servidor central para el tráfico hacia internet. -- To UNSUBSCRIBE, email to debian-user-spanish-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/5331418d.6010...@gmail.com
Re: Inexplicable broken pipe entre dos subredes bajo mismo LAN
El 25/03/2014 9:35, Roberto escribió: El 24/03/2014 11:34, Francesc Guitart escribió: El 23/03/2014 10:32, Roberto escribió: Administro una red LAN donde siempre que puedo aplico Debian a la ecuación. El servidor que hace de gateway hace de firewall y registra log determinado trafico. Tiene dos interfaces br0 y br0:1 con 192.168.0.2 y 192.168.2.1. Uso un alias por no disponer de dos tarjetas de red, y uso br0 porque en dicho equipo uso VirtualBox. Tengo un servidor con jenkins(Debian) en la 192.168.0.8 el cual conecta por ssh a dos nodos esclavos, 192.168.2.19(Centos) y 192.168.2.21(Ubuntu). El primero tiene como gateway a 192.168.0.2 y el segundo a 192.168.2.1. Dices que el CentOS tiene IP 192.168.2.19 y gateway 192.168.0.2. ¿La gateway del CentOS no debería ser también 192.168.2.1 si está en dicha red? Con Ubuntu ningún problema con el trafico de red pero con Centos aleatoriamente ocurre un broken pipe de una sesión ssh, lo único que se me ocurrió despues de días dando cabezasos es pasar ese nodo Centos a la subred con ip 192.168.0.9 y todo arreglado. Abajo pegaré el log de un broken pipe. Como dice Camaleón me parece raro que sea por culpa de una ruta pero creo que la gateway del CentOS no es correcta. Yo la cambiaría a ver si el error se reproduce. ¿Hay alguna idea para en esta situación se produzca un broken pipe? Y he configurado Centos para no usar Gssapi, DenyDNS, y lo típico en cliente para evitar un time out. jenkins@192.168.2.19's password: debug3: packet_send2: adding 64 (len 60 padlen 4 extra_pad 64) (...) El gateway del Centos es 192.168.2.1, al igual que el nodo de Ubuntu. Pero la maquina Debian que los virtualiza lleva un bridged br0 con ip en la subred 192.168.0.7 ,pero vamos un bridged deja pasar todo el trafico de sus maquinas virtualizadas. Lo siento, no te entendí. Ahora lo tengo más claro. ¿Has mirado los logs en el servidor que recibe la conexión? Si no hay ninguna pista incrementa el nivel de log poniendo VERBOSE o DEBUG o DEBUG1... (hasta DEBUG3) en el parámetro LogLevel de /etc/sshd_config. Reinicia sshd, intenta de nuevo la conexión y envía la salida de los logs si tienes dudas sobre algo. Saludos. -- Francesc Guitart -- To UNSUBSCRIBE, email to debian-user-spanish-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/5331443c.2030...@gmx.com
Re: Inexplicable broken pipe entre dos subredes bajo mismo LAN
El 23/03/2014 10:32, Roberto escribió: Administro una red LAN donde siempre que puedo aplico Debian a la ecuación. El servidor que hace de gateway hace de firewall y registra log determinado trafico. Tiene dos interfaces br0 y br0:1 con 192.168.0.2 y 192.168.2.1. Uso un alias por no disponer de dos tarjetas de red, y uso br0 porque en dicho equipo uso VirtualBox. Tengo un servidor con jenkins(Debian) en la 192.168.0.8 el cual conecta por ssh a dos nodos esclavos, 192.168.2.19(Centos) y 192.168.2.21(Ubuntu). El primero tiene como gateway a 192.168.0.2 y el segundo a 192.168.2.1. Dices que el CentOS tiene IP 192.168.2.19 y gateway 192.168.0.2. ¿La gateway del CentOS no debería ser también 192.168.2.1 si está en dicha red? Con Ubuntu ningún problema con el trafico de red pero con Centos aleatoriamente ocurre un broken pipe de una sesión ssh, lo único que se me ocurrió despues de días dando cabezasos es pasar ese nodo Centos a la subred con ip 192.168.0.9 y todo arreglado. Abajo pegaré el log de un broken pipe. Como dice Camaleón me parece raro que sea por culpa de una ruta pero creo que la gateway del CentOS no es correcta. Yo la cambiaría a ver si el error se reproduce. ¿Hay alguna idea para en esta situación se produzca un broken pipe? Y he configurado Centos para no usar Gssapi, DenyDNS, y lo típico en cliente para evitar un time out. jenkins@192.168.2.19's password: debug3: packet_send2: adding 64 (len 60 padlen 4 extra_pad 64) debug2: we sent a password packet, wait for reply debug1: Authentication succeeded (password). Authenticated to 192.168.2.19 ([192.168.2.19]:22). debug1: channel 0: new [client-session] debug3: ssh_session2_open: channel_new: 0 debug2: channel 0: send open debug1: Requesting no-more-sessi...@openssh.com debug1: Entering interactive session. debug2: callback start debug2: client_session2_setup: id 0 debug2: fd 3 setting TCP_NODELAY debug3: packet_set_tos: set IP_TOS 0x10 debug2: channel 0: request pty-req confirm 1 debug1: Sending environment. debug3: Ignored env SHELL debug3: Ignored env TERM debug3: Ignored env USER debug3: Ignored env LS_COLORS debug3: Ignored env MAIL debug3: Ignored env PATH debug3: Ignored env PWD debug1: Sending env LANG = es_ES.UTF-8 debug2: channel 0: request env confirm 0 debug3: Ignored env SHLVL debug3: Ignored env HOME debug3: Ignored env LOGNAME debug3: Ignored env _ debug2: channel 0: request shell confirm 1 debug2: callback done debug2: channel 0: open confirm rwindow 0 rmax 32768 Write failed: Broken pipe -- Francesc Guitart -- To UNSUBSCRIBE, email to debian-user-spanish-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/53300a49.7040...@gmx.com
Inexplicable broken pipe entre dos subredes bajo mismo LAN
Administro una red LAN donde siempre que puedo aplico Debian a la ecuación. El servidor que hace de gateway hace de firewall y registra log determinado trafico. Tiene dos interfaces br0 y br0:1 con 192.168.0.2 y 192.168.2.1. Uso un alias por no disponer de dos tarjetas de red, y uso br0 porque en dicho equipo uso VirtualBox. Tengo un servidor con jenkins(Debian) en la 192.168.0.8 el cual conecta por ssh a dos nodos esclavos, 192.168.2.19(Centos) y 192.168.2.21(Ubuntu). El primero tiene como gateway a 192.168.0.2 y el segundo a 192.168.2.1. Con Ubuntu ningún problema con el trafico de red pero con Centos aleatoriamente ocurre un broken pipe de una sesión ssh, lo único que se me ocurrió despues de días dando cabezasos es pasar ese nodo Centos a la subred con ip 192.168.0.9 y todo arreglado. Abajo pegaré el log de un broken pipe. ¿Hay alguna idea para en esta situación se produzca un broken pipe? Y he configurado Centos para no usar Gssapi, DenyDNS, y lo típico en cliente para evitar un time out. jenkins@192.168.2.19's password: debug3: packet_send2: adding 64 (len 60 padlen 4 extra_pad 64) debug2: we sent a password packet, wait for reply debug1: Authentication succeeded (password). Authenticated to 192.168.2.19 ([192.168.2.19]:22). debug1: channel 0: new [client-session] debug3: ssh_session2_open: channel_new: 0 debug2: channel 0: send open debug1: Requesting no-more-sessi...@openssh.com debug1: Entering interactive session. debug2: callback start debug2: client_session2_setup: id 0 debug2: fd 3 setting TCP_NODELAY debug3: packet_set_tos: set IP_TOS 0x10 debug2: channel 0: request pty-req confirm 1 debug1: Sending environment. debug3: Ignored env SHELL debug3: Ignored env TERM debug3: Ignored env USER debug3: Ignored env LS_COLORS debug3: Ignored env MAIL debug3: Ignored env PATH debug3: Ignored env PWD debug1: Sending env LANG = es_ES.UTF-8 debug2: channel 0: request env confirm 0 debug3: Ignored env SHLVL debug3: Ignored env HOME debug3: Ignored env LOGNAME debug3: Ignored env _ debug2: channel 0: request shell confirm 1 debug2: callback done debug2: channel 0: open confirm rwindow 0 rmax 32768 Write failed: Broken pipe -- To UNSUBSCRIBE, email to debian-user-spanish-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/532eaa40.5060...@gmail.com
Re: Inexplicable broken pipe entre dos subredes bajo mismo LAN
El Sun, 23 Mar 2014 10:32:48 +0100, Roberto escribió: (...) Tengo un servidor con jenkins(Debian) en la 192.168.0.8 el cual conecta por ssh a dos nodos esclavos, 192.168.2.19(Centos) y 192.168.2.21(Ubuntu). El primero tiene como gateway a 192.168.0.2 y el segundo a 192.168.2.1. Con Ubuntu ningún problema con el trafico de red pero con Centos aleatoriamente ocurre un broken pipe de una sesión ssh, lo único que se me ocurrió despues de días dando cabezasos es pasar ese nodo Centos a la subred con ip 192.168.0.9 y todo arreglado. Abajo pegaré el log de un broken pipe. (...) El hecho de cambiar la red del cliente (CentOS) y que funcione es desde luego significativo pero como dices que se trata de un error aleatorio en principio no me centraría en un problema de rutas (dado que el resto de tráfico de red dices que funciona), sino en el error en sí mismo, es decir, investiga qué situaciones pueden generar ese timeout de SSH: https://www.google.com/webhp?complete=0hl=en#complete=0hl=enq=ssh+%22Write+failed:+Broken+pipe%22 Saludos, -- Camaleón -- To UNSUBSCRIBE, email to debian-user-spanish-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/pan.2014.03.23.14.19...@gmail.com
Re: [Bulk] Re: ssh error - Write failed: Broken pipe
On 24/05/12 04:24 AM, elbbit wrote: On 23/05/12 22:45, Gary Dale wrote: On 23/05/12 04:53 PM, elbbit wrote: On your server put this line in your sshd_config after Port 22: Port 443 Do a service ssh restart or /etc/init.d/ssh restart. Connect using: ssh -p 443 user@host That has the same issue as the keepalive - it presumes that the connection is at fault when the problem only seems to happen when I'm doing large amounts of data movement on the server (NOT over the connection). Are you saying that if you log in by SSH then issue a command like cp large_file4gb large_copy4gb that even though the command produces no output you then get broken pipe? If so the connection is likely being terminated because it's idle (check your internet provider). The keepalive should prevent this. If you log in and do a ls -R / do you get the output or broken pipe? If you get broken pipe before the command completes this again suggests (to me) that you are subject to packet inspection. For me, putting sshd on different ports helps here like ports 22, 81, 443, 2022, 3389 and 20222. Some 3G providers here (UK) block port 22 altogether. One works best on port 81, on others 443 seems to work most of the time. Some network admins prefer to administer Windows boxes and (thus) leave port 3389 untouched. YMMV. Either way, in my experience broken pipe is usually a network problem. Fascinating. The keep-alive seems to prevent the error, which is occurring at the end of whatever command I executed on the server. This would be when the remote shell is returning to the command prompt. For example, if I mount usb drive on the remote server then copy some files to it, the pipe breaks after the copy completes. If I put the sequence in a script, complete it with a umount command at the end, the copy completes and the device is not mounted when the pipe breaks. This would suggest that it's the sending of data (the shell prompt) from the remote machine that is triggering the broken pipe. This is confirmed because I can leave the remote session sitting for hours before coming back to enter another command without problems. It's just when it's the remote server initiating data transfer after an extended period of inactivity that I get a broken pipe. Clearly the keep-alive is helping by preventing an extended period of inactivity. However, I'm puzzled by why the timeout only seems to operate against the server and not against my workstation. The remote servers are running Squeeze in all cases, while my workstation is running Wheezy. Has something changed between the two releases? -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4fbfd299.3030...@rogers.com
Re: [Bulk] Re: ssh error - Write failed: Broken pipe
On 23/05/12 22:45, Gary Dale wrote: On 23/05/12 04:53 PM, elbbit wrote: On your server put this line in your sshd_config after Port 22: Port 443 Do a service ssh restart or /etc/init.d/ssh restart. Connect using: ssh -p 443 user@host That has the same issue as the keepalive - it presumes that the connection is at fault when the problem only seems to happen when I'm doing large amounts of data movement on the server (NOT over the connection). Are you saying that if you log in by SSH then issue a command like cp large_file4gb large_copy4gb that even though the command produces no output you then get broken pipe? If so the connection is likely being terminated because it's idle (check your internet provider). The keepalive should prevent this. If you log in and do a ls -R / do you get the output or broken pipe? If you get broken pipe before the command completes this again suggests (to me) that you are subject to packet inspection. For me, putting sshd on different ports helps here like ports 22, 81, 443, 2022, 3389 and 20222. Some 3G providers here (UK) block port 22 altogether. One works best on port 81, on others 443 seems to work most of the time. Some network admins prefer to administer Windows boxes and (thus) leave port 3389 untouched. YMMV. Either way, in my experience broken pipe is usually a network problem. -- elb...@gmail.com -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4fbdf036.7080...@gmail.com
Re: ssh error - Write failed: Broken pipe
Hi - entering this thread late... On Wed, May 23, 2012 at 07:46:29PM +0100, Gary Dale wrote: I'm connecting at various times to different Debian/Squeeze servers from my Debian/Wheezy workstation using ssh. No matter which server I connect to, I find that if I move a lot of data on the remote server, I get kicked out of my ssh session with a Write failed: Broken pipe error. I've seen that before and it took us ages to narrow down - in our case (admittedly your case may be different), it was the combination of network interface MTU and packet fragmentation. Reducing the MTU on our local network interface (on the server which runs the SSH client) to 1450 (rather than the default 1500) solved it consistently: sudo ifconfig eth0 mtu 1450 (to make the effect permanent, tweak /etc/network/interfaces and/or network manager config) Unfortunately we never found the root cause of the problem: Packet fragmentation is normally handled transparently by the TCP layer and there were no VPNs or VLANs involved. This doesn't happen immediately. It usually occurs after some period of time, during which time the cp or rsync command is moving data from one location to another (gigabytes typically). So far as I can tell, the operation finished successfully. Interesting. This is repeatable 100% of the time. typical commands: cp /tmp/backups/* /media/dvd rsync /tmp/backups/file /media/dvd/file Hm. These commands would not use the network - unless /tmp or /media happen to be NFS mounted, ISCSI-backed or similar - and neither involves ssh... I have even encountered it with if cmp /media/backupmedia/backup-set /home/shares/backup/backup-set; then echo OK; fi Hm.. this example does not involve ssh at all!? And (apparantly) not network either? although this sometimes works. I think it depends on the exact size of the file. The cmp operation seems to not trigger it as fast as rsync or cp. If I'm not moving large amounts of data on the server, the ssh session is stable and will sit for as long as I want. I only get disconnected when I'm moving data around on the remote machine. That is consistent with the symptoms we saw before decreasing the MTU. Hope this helps -- Karl E. Jorgensen IT Operations -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120524113910.GQ17862@hawking
Re: ssh error - Write failed: Broken pipe
On Wed, May 23, 2012 at 08:25:20PM +0100, Gary Dale wrote: On 23/05/12 03:07 PM, elbbit wrote: On 23/05/12 19:46, Gary Dale wrote: This is repeatable 100% of the time. I get this error on my ISP. I have found they are inspecting the internet packets and terminating if there is too little back-and-forth traffic. My ISP is a mobile phone network over 3G so they are probably trying to keep the bandwidth for phones only. I do -o ServerAliveInterval=5 which has my ssh client ping the server every 5 seconds to keep it alive. To summarise, I get rid of broken pipe errors by using this command line ssh -o ServerAliveInterval=5 u...@server.com Hope this helps, elbbit I can try it, but as I said, I have a stable connection when I'm not doing anything. I can leave an ssh shell open overnight and it will be there in the morning. It's only when I'm moving bits around on the remote machine that I get disconnected. ah. I misunderstood your scenario in my previous mail... sorry. So the ssh breakage is (somehow) triggered by moving bits around on the remote machine only - not by moving lots of bits through the SSH connection. Anything in the logs on the remote machine? these files may contain relevant stuff: /var/log/syslog /var/log/daemon.log /var/log/kern.log .. basically all of them... look for files in /var/log which were touched around the time (or later) of the problem. Hope this helps -- Karl E. Jorgensen IT Operations -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120524114601.GR17862@hawking
Re: ssh error - Write failed: Broken pipe
Karl E. Jorgensen karl.jorgen...@nice.com wrote: I've seen that before and it took us ages to narrow down - in our case (admittedly your case may be different), it was the combination of network interface MTU and packet fragmentation. I was just going to suggest exactly the same thing. And it took us ages to track down, too! Chris -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/a6v199x2pt@news.roaima.co.uk
ssh error - Write failed: Broken pipe
I'm connecting at various times to different Debian/Squeeze servers from my Debian/Wheezy workstation using ssh. No matter which server I connect to, I find that if I move a lot of data on the remote server, I get kicked out of my ssh session with a Write failed: Broken pipe error. This doesn't happen immediately. It usually occurs after some period of time, during which time the cp or rsync command is moving data from one location to another (gigabytes typically). So far as I can tell, the operation finished successfully. This is repeatable 100% of the time. typical commands: cp /tmp/backups/* /media/dvd rsync /tmp/backups/file /media/dvd/file (I'd be doing this to fix a faulty file copy onto Blu-Ray media, for example) I have even encountered it with if cmp /media/backupmedia/backup-set /home/shares/backup/backup-set; then echo OK; fi although this sometimes works. I think it depends on the exact size of the file. The cmp operation seems to not trigger it as fast as rsync or cp. If I'm not moving large amounts of data on the server, the ssh session is stable and will sit for as long as I want. I only get disconnected when I'm moving data around on the remote machine. Any ideas? -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4fbd3085.5030...@rogers.com
Re: ssh error - Write failed: Broken pipe
On 23/05/12 19:46, Gary Dale wrote: This is repeatable 100% of the time. I get this error on my ISP. I have found they are inspecting the internet packets and terminating if there is too little back-and-forth traffic. My ISP is a mobile phone network over 3G so they are probably trying to keep the bandwidth for phones only. I do -o ServerAliveInterval=5 which has my ssh client ping the server every 5 seconds to keep it alive. To summarise, I get rid of broken pipe errors by using this command line ssh -o ServerAliveInterval=5 u...@server.com Hope this helps, elbbit -- elb...@gmail.com -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4fbd3582.6060...@gmail.com
Re: ssh error - Write failed: Broken pipe
On 23/05/12 03:07 PM, elbbit wrote: On 23/05/12 19:46, Gary Dale wrote: This is repeatable 100% of the time. I get this error on my ISP. I have found they are inspecting the internet packets and terminating if there is too little back-and-forth traffic. My ISP is a mobile phone network over 3G so they are probably trying to keep the bandwidth for phones only. I do -o ServerAliveInterval=5 which has my ssh client ping the server every 5 seconds to keep it alive. To summarise, I get rid of broken pipe errors by using this command line ssh -o ServerAliveInterval=5 u...@server.com Hope this helps, elbbit I can try it, but as I said, I have a stable connection when I'm not doing anything. I can leave an ssh shell open overnight and it will be there in the morning. It's only when I'm moving bits around on the remote machine that I get disconnected. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4fbd39a0.3050...@rogers.com
Re: ssh error - Write failed: Broken pipe
On 23/05/12 20:25, Gary Dale wrote: On 23/05/12 03:07 PM, elbbit wrote: On 23/05/12 19:46, Gary Dale wrote: This is repeatable 100% of the time. I get this error on my ISP. I have found they are inspecting the internet packets and terminating if there is too little back-and-forth traffic. My ISP is a mobile phone network over 3G so they are probably trying to keep the bandwidth for phones only. I do -o ServerAliveInterval=5 which has my ssh client ping the server every 5 seconds to keep it alive. To summarise, I get rid of broken pipe errors by using this command line ssh -o ServerAliveInterval=5 u...@server.com Hope this helps, elbbit I can try it, but as I said, I have a stable connection when I'm not doing anything. I can leave an ssh shell open overnight and it will be there in the morning. It's only when I'm moving bits around on the remote machine that I get disconnected. In that case you may have the same problem I did some months ago. I'm on T-Mobile here and they like to packet inspect the TCP sessions. I found by putting all SSH traffic through port 443 I subverted their system -- they don't inspect encrypted traffic for which port 443 is the default for secure http. On your server put this line in your sshd_config after Port 22: Port 443 Do a service ssh restart or /etc/init.d/ssh restart. Connect using: ssh -p 443 user@host See if that helps. elbbit -- elb...@gmail.com -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4fbd4e46.4080...@gmail.com
Re: [Bulk] Re: ssh error - Write failed: Broken pipe
On 23/05/12 04:53 PM, elbbit wrote: On 23/05/12 20:25, Gary Dale wrote: On 23/05/12 03:07 PM, elbbit wrote: On 23/05/12 19:46, Gary Dale wrote: This is repeatable 100% of the time. I get this error on my ISP. I have found they are inspecting the internet packets and terminating if there is too little back-and-forth traffic. My ISP is a mobile phone network over 3G so they are probably trying to keep the bandwidth for phones only. I do -o ServerAliveInterval=5 which has my ssh client ping the server every 5 seconds to keep it alive. To summarise, I get rid of broken pipe errors by using this command line ssh -o ServerAliveInterval=5 u...@server.com Hope this helps, elbbit I can try it, but as I said, I have a stable connection when I'm not doing anything. I can leave an ssh shell open overnight and it will be there in the morning. It's only when I'm moving bits around on the remote machine that I get disconnected. In that case you may have the same problem I did some months ago. I'm on T-Mobile here and they like to packet inspect the TCP sessions. I found by putting all SSH traffic through port 443 I subverted their system -- they don't inspect encrypted traffic for which port 443 is the default for secure http. On your server put this line in your sshd_config after Port 22: Port 443 Do a service ssh restart or /etc/init.d/ssh restart. Connect using: ssh -p 443 user@host See if that helps. elbbit That has the same issue as the keepalive - it presumes that the connection is at fault when the problem only seems to happen when I'm doing large amounts of data movement on the server (NOT over the connection). -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4fbd5a89.3060...@rogers.com
Re: ssh error - Write failed: Broken pipe
On 23/05/12 03:25 PM, Gary Dale wrote: On 23/05/12 03:07 PM, elbbit wrote: On 23/05/12 19:46, Gary Dale wrote: This is repeatable 100% of the time. I get this error on my ISP. I have found they are inspecting the internet packets and terminating if there is too little back-and-forth traffic. My ISP is a mobile phone network over 3G so they are probably trying to keep the bandwidth for phones only. I do -o ServerAliveInterval=5 which has my ssh client ping the server every 5 seconds to keep it alive. To summarise, I get rid of broken pipe errors by using this command line ssh -o ServerAliveInterval=5 u...@server.com Hope this helps, elbbit I can try it, but as I said, I have a stable connection when I'm not doing anything. I can leave an ssh shell open overnight and it will be there in the morning. It's only when I'm moving bits around on the remote machine that I get disconnected. Strange. When I turned on the keepalive, the cmp completed faster / I got the remote prompt back faster with no broken pipe. The stable connection could be because the operation returned faster or it could be that the keepalive fixed the problem. I'll have to check some more. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4fbd5bb0.5010...@rogers.com
Broken Pipe / self-DOS
Hallo! Hab vorige Woche ein apt-get upgrade gemacht; seither habe ich bei der aktuellen testing folgendes Problem: Wenn man in einer shell so etwas schreibt wie: [EMAIL PROTECTED]:~$ more ~ und dann zweimal TAB drückt, schreibt er noch brav hin Display all 201 possibilities? (y or n) aber ab dann ist die shell kaputt. kein ls, kein ps, kein xterm, keine gar nichts. Guckst du. [EMAIL PROTECTED]:~$ ls [EMAIL PROTECTED]:~$ ps [EMAIL PROTECTED]:~$ more /etc/fstab [EMAIL PROTECTED]:~$ cat /proc/cpuinfo [EMAIL PROTECTED]:~$ mount [EMAIL PROTECTED]:~$ ? [EMAIL PROTECTED]:~$ blödsinn [EMAIL PROTECTED]:~$ bash, csh, ksh, sh, tcsh - alle kaputt. Die tcsh, liefert zumindest noch ne Fehlermeldung pro versuchten Aufruf: broken pipe. Ich kann leider nicht erruieren in welchem Paket das Problem auftaucht. ldap, pam, login, tty, libnss - keine Ahnung. Wo könnt ich diesen fiesen Bug denn reporten?? Achso, für die Zweifler noch: das Problem hab ich nicht nur in unseren Systemen gefunden sondern auch auf alles testing Rechner vom CIP Pool. Grüße, MIRKO
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Iñaki wrote: Servidor de correo a medida? De que estas hablando? Cuanto tiempo, esfuerzo y trabajo te lleva implementar y dejar funcionando con mantenimiento cercano a cero eso que te plantee? Me explico: una cosa es dejar el servidor perfectamente funcional, seguro y con todas las posibilidades del mundo y otra cosa es el grado de dependencia que quien te lo ha contratado tendrá contigo. Estoy convencido de que les provees mecanismos sencillos para que puedan hacer algo tan banal como crear crear/eliminar/modificar cuentas de correo sin tener que recurrir a ti, ¿verdad? Si. Pero ¿y si cambian de dominio? ¿pueden ellos mismos hacer las modificaciones necesarias en el servidor de correo (y todos sus componentes) para que siga en funcionamiento? Si. lo pregunto por curiosidad, de verdad, es una cosa que tengo en la cabeza desde hace tiempo. - -- Ricardo A.Frydman Consultor en Tecnología Open Source - Administrador de Sistemas jabber: [EMAIL PROTECTED] - http://www.eureka-linux.com.ar SIP # 1-747-667-9534 -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iD8DBQFEDCmukw12RhFuGy4RAk0wAKCORoZoniukGJNUjAA3dM5xiSHKsgCgg6EL Yq1Uz+/3YmeWsiQc2BsfdVw= =0GSk -END PGP SIGNATURE-
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
El Lunes, 6 de Marzo de 2006 13:23, Ricardo Frydman Eureka! escribió: Iñaki wrote: Servidor de correo a medida? De que estas hablando? Cuanto tiempo, esfuerzo y trabajo te lleva implementar y dejar funcionando con mantenimiento cercano a cero eso que te plantee? Me explico: una cosa es dejar el servidor perfectamente funcional, seguro y con todas las posibilidades del mundo y otra cosa es el grado de dependencia que quien te lo ha contratado tendrá contigo. Estoy convencido de que les provees mecanismos sencillos para que puedan hacer algo tan banal como crear crear/eliminar/modificar cuentas de correo sin tener que recurrir a ti, ¿verdad? Si. Pero ¿y si cambian de dominio? ¿pueden ellos mismos hacer las modificaciones necesarias en el servidor de correo (y todos sus componentes) para que siga en funcionamiento? Si. Pues entonces felicidades por un trabajo bien hecho ;) -- y hasta aquí puedo leer...
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Antonio Galicia wrote: On 3/3/06, Ricardo Frydman Eureka! [EMAIL PROTECTED] wrote: PD: Aprovecho para decir que en Linux no existe ningún servidor de correo como solución final (SMTP, POP, IMAP, autenticación, reglas, interfaz web de mantenimiento y configuración con Logs, gráficas y demás reportes, todo ello integrado). Por supuesto! Y Dios permita que esto nunca llegue ¿? ¿Razón? small is beatifull Filosofia *nix: hacer una sola cosa bien y eficientemente. Filosofia GNU/FLOSS: Libertad de eleccion/variedad de gustos sabores e implementacionesalcanzan? Luego pediran una distribucion/escritorio totalmente integrado y asi reinventaran las ventanas (no la rueda)...pero libres ¿No es lo que hace (casí) Ubuntú? No. En todo caso es lo que pretende KDE y/o Gnome (por cierto es Ubuntu, sin acento) No conozco CommuniGate, pero mis clientes quedan encantados con: - - postfix - - cyrus (solo IMAP) - - usermin y webmin - - horde (con imp, gollem y turba) - - LDAP (Generalmente, son desertores de las ventanitas /hartos/ de problemas) Perfecto, espero que tienes documentado (un HOWTO) de como se hace todo esto Casi ... ya falta poco. y puedes ponerlo al servicio de la comunidad como muestra de agradecimiento al trabajo en que basas el tuyo propio. mmm no, lo pondre como manera de compartir. La manera de agradecer que tengo es solicitar a mis clientes que soporten ($$$) a las empresas cuyas proyectos estan involucrados. Pero eso no lo hacemos publico ;) Otra es ayudar en algunas listas de correo. Por cierto...tienes toneladas de documentacion al respecto, sabias? Y lo vas a hacer público en los próximos dias, ¿cierto? No. Apenas tenga tiempo disponible, que no es lo mismo. ultimamente estoy muy ocupado explicando a la gente porque hay reglas en las listas de correo. usar servidores de correo propietarios salvo que se pongan ellos mismos a diseñarse su Postfix con LDAP, Cyrus y demás... Para que diseñar uno propio si ya existe uno? Y se llama ... ¿? Te puse una combinacion posible alli arriba, releela. Yo detestaria manejar eso por web ;) ¿Y qué fue del usermin y webmin? Los detesto, por supuesto, pero los que lo usan lo adoran. del CommuniGate pinchas en una pestañita y ves en tiempo real los correos que se están enviando, recibiendo, gráficas, estadísticas y demás. Eso si que es util! Ug... debe serlo para alguien y está dispuesto a pagar por tenerlo. No te imaginas las cosas inutiles que me han pagado por hacer! ¿No ves una oportunidad de negocio? Si tienes armado todo el proceso que le ofreces a tus clientes se puede agregar este tipo de estadísticas y venderlo como un paquete completo. ¿No? Esas cosas se agradecen y a veces son imprescindibles. Como decian en los tiempos de mi abuela: la juventud esta perdida. No hagas caso, mi agüela decía lo mismo... y la abuela de ella otro tanto. No se si se han equivocado tanto en muchas ocasiones... -- Saludos, Toño http://agc.com.mx/antoniogc/blog - -- Ricardo A.Frydman Consultor en Tecnología Open Source - Administrador de Sistemas jabber: [EMAIL PROTECTED] - http://www.eureka-linux.com.ar SIP # 1-747-667-9534 -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iD8DBQFECaIlkw12RhFuGy4RAshJAJ4rGQjrToor5DSRM4WnLNyi3GohzQCeNJWz RXmhHOJa1bpYA3fnaxav6W0= =SY0F -END PGP SIGNATURE-
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Iñaki wrote: El Viernes, 3 de Marzo de 2006 16:25, Ricardo Frydman Eureka! escribió: Iñaki wrote: PD: Aprovecho para decir que en Linux no existe ningún servidor de correo como solución final (SMTP, POP, IMAP, autenticación, reglas, interfaz web de mantenimiento y configuración con Logs, gráficas y demás reportes, todo ello integrado). Por supuesto! Y Dios permita que esto nunca llegue Entonces Linux sólo será para administradores experimentados que se montan ellos mismo su SMTP, POP, IMAP y absolutamente todo, y que sólo dicho administrador sea capaz de meter mano a lo que ha creado. Escuchaste hablar de los puntos medios? El empresario que tiene una oficina con 4 empleados y se atreve con cosas de Windows más gráficas que se olvide de Linux, porque? De hecho son quienes mas se atreven a estos cambios... ese no es de los buenos, que use soluciones integradas y finales de Microsoft o que pague a un consultor para que le diseñe un servidor de correo a medida. Servidor de correo a medida? De que estas hablando? Cuanto tiempo, esfuerzo y trabajo te lleva implementar y dejar funcionando con mantenimiento cercano a cero eso que te plantee? Yo no creo que haya ningún problema en que Linux como servidor se haga más fácil y accesible a gente que si no se acaba de acercar es porque teme a la consola. Porque confundes las cosas? Estas seguro que leiste en mi correo lo que yo escribi? La gente que esta migrando a Linux y Software Libre en empresas medianas y grandes como en minusculas son unos idiotas hipnotizados? Que yo sepa sólo existe Kolab (www.kolab.org) que ofrece SMTP, POP, IMAP, autenticación, agenda LDAP, groupware y un bonito interfaz web para administrar usuarios y demás. Tiene muy buena pinta pero de momento no tiene ni la mitad que el CommuniGate (he configurado ambos). No conozco CommuniGate, pero mis clientes quedan encantados con: - postfix - cyrus (solo IMAP) - usermin y webmin - horde (con imp, gollem y turba) - LDAP Por supuesto, pero que quede bien claro que sólo tú serás capaz de meter mano a ese servidor que has diseñado a tu gusto, Te equivocas de punta a punta: yo implemento configuro explico y en general no me necesitan mas. y que seguro que es estupendo, No he recibido quejas. pero si algún día necesitan asistencia y tú no puedes atenderles tendrán que gastarse un dineral en contratar a un consultor para que investigue cómo funciona el invento. Iñaki: me sorprende muchisimo como hablas tanto y de tan mala manera de algo que ni siquiera has tenido la mas minima posibilidad de ver ni remotamente. No pienso entablar una discusion contigo (ni con nadie) de este nivel y mucho menos publicamete, solo decirte que si necesitas soporte de CommuniBillGates, te contactes con la empresa y no la pidas por aqui. Punto. Por cierto, kolab.org es precisamente: - postfix - cyrus (POP e IMAP) - OpenLDAP - Interfaz web de administración - Clientes como horde o Kontact y conectores para Mozilla y propietarios para Outlook (para el tema de groupware). Lo pruebas y nos cuentas. (Generalmente, son desertores de las ventanitas /hartos/ de problemas) En realidad yo sólo hablo de un interfaz web de administración centralizada, nada más. No me parece tan descabellado que empresas que apuesten por Linux tengan que Descabellado, no. Solo inutil. usar servidores de correo propietarios salvo que se pongan ellos mismos a diseñarse su Postfix con LDAP, Cyrus y demás... Para que diseñar uno propio si ya existe uno? ¿A cuál te refieres? y luego a administrarlo por consola (estoy hablando de un caso de un servidor con más de 3000 usuarios, a mí no me gustaría llevar su administración por consola). Yo detestaria manejar eso por web ;) Hombre, pero una web es accesible desde cualquier sitio, navegador y SO, y si te enseña cositas gráficamente de forma más agradable que una fría consola... ¿qué tiene de malo? Quien dijo que fuera malo? Solo dije que yo lo detestaria! Sigues leyendo lo que tienes ganas de leer. En el interfaz web del CommuniGate pinchas en una pestañita y ves en tiempo real los correos que se están enviando, recibiendo, gráficas, estadísticas y demás. Eso si que es util! Y hasta divertido ;) Esas cosas se agradecen y a veces son imprescindibles. Como decian en los tiempos de mi abuela: la juventud esta perdida. Tranquilo, esa frase la diremos todos algún día ;) - -- Ricardo A.Frydman Consultor en Tecnología Open Source - Administrador de Sistemas jabber: [EMAIL PROTECTED] - http://www.eureka-linux.com.ar SIP # 1-747-667-9534 -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iD8DBQFECaPdkw12RhFuGy4RAor1AJ9aOHXsCqbUD6EDLWnQiiqnzHpYSACZAYsc OZAudI97RfB3kdQoz+Ae3xk= =YVOB -END PGP SIGNATURE-
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
On 3/4/06, Ricardo Frydman Eureka! [EMAIL PROTECTED] wrote: small is beatifull Supongo entonces que el proyecto gnome te parece una aberración. Y mejor ni hablar de mono. Filosofia *nix: hacer una sola cosa bien y eficientemente. Filosofia GNU/FLOSS: Libertad de eleccion/variedad de gustos sabores e implementacionesalcanzan? Tienes razón y no tiene caso saguir con esto. -- Saludos, Toño http://agc.com.mx/antoniogc/blog
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
El Sábado, 4 de Marzo de 2006 15:27, Ricardo Frydman Eureka! escribió: ese no es de los buenos, que use soluciones integradas y finales de Microsoft o que pague a un consultor para que le diseñe un servidor de correo a medida. Servidor de correo a medida? De que estas hablando? Cuanto tiempo, esfuerzo y trabajo te lleva implementar y dejar funcionando con mantenimiento cercano a cero eso que te plantee? Me explico: una cosa es dejar el servidor perfectamente funcional, seguro y con todas las posibilidades del mundo y otra cosa es el grado de dependencia que quien te lo ha contratado tendrá contigo. Estoy convencido de que les provees mecanismos sencillos para que puedan hacer algo tan banal como crear crear/eliminar/modificar cuentas de correo sin tener que recurrir a ti, ¿verdad? Pero ¿y si cambian de dominio? ¿pueden ellos mismos hacer las modificaciones necesarias en el servidor de correo (y todos sus componentes) para que siga en funcionamiento? lo pregunto por curiosidad, de verdad, es una cosa que tengo en la cabeza desde hace tiempo. No conozco CommuniGate, pero mis clientes quedan encantados con: - postfix - cyrus (solo IMAP) - usermin y webmin - horde (con imp, gollem y turba) - LDAP Por supuesto, pero que quede bien claro que sólo tú serás capaz de meter mano a ese servidor que has diseñado a tu gusto, Te equivocas de punta a punta: yo implemento configuro explico y en general no me necesitan mas. pero si algún día necesitan asistencia y tú no puedes atenderles tendrán que gastarse un dineral en contratar a un consultor para que investigue cómo funciona el invento. Iñaki: me sorprende muchisimo como hablas tanto y de tan mala manera de algo que ni siquiera has tenido la mas minima posibilidad de ver ni remotamente. Ricardo, he leído lo que dije y efectivamente parece que critico algo que no conozco, no era mi intención, acepta mis disculpas. No pienso entablar una discusion contigo (ni con nadie) de este nivel y mucho menos publicamete, solo decirte que si necesitas soporte de CommuniBillGates, te contactes con la empresa y no la pidas por aqui. Punto. CommuniGate no es de Microsoft, sino de Stallker, lo cual tampoco viene al caso. No soy ningún defensor de este servidor de correo, simplemente me ha tocado hacer consultoría para una empresa que lo usa y me tengo que pelear con él. Pero en una cosa te equivocas: si lees mi correo original verás que en ningún caso pido ayuda sobre el servidor de correo propietario, sino sobre el error de sed, y lo que me interesa es saber en qué contexto se puede producir el error de tubería rota para poder investigarlo, sólo eso. Por cierto, kolab.org es precisamente: - postfix - cyrus (POP e IMAP) - OpenLDAP - Interfaz web de administración - Clientes como horde o Kontact y conectores para Mozilla y propietarios para Outlook (para el tema de groupware). Lo pruebas y nos cuentas. Creo que ya lo había dicho: lo he probado e instalado y puedo hablar muy bien de él. Lo que me sorprende es precisamente que sea la única solución final como servidor de correo GPL que he encontrado. Saludos. -- y hasta aquí puedo leer...
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Iñaki wrote:
El Jueves, 2 de Marzo de 2006 20:03, Antonio Galicia escribió:
On 3/2/06, Iñaki [EMAIL PROTECTED] wrote:
Un problema similar sucede si uso awk en vez de sed:
-
--- cat $correo_original | awk
'ok{print}/^\s*$/{ok=1}' | formail -X $cabecera_original
-
---
Similar ... ¿tienes el texto del error?
Sí, usando esta expresión de sed:
cat $correo_original | sed '1,/^$/ d' | formail -X $cabecera_original
el error sería éste:
11:23:54.64 1 ENQUEUERRULES [9570439] rule(myRule) action #1:
execute stderr: sed: couldn't write 72 items to {standard output}:
Broken pipe: error de escritura: Tuberia rota
--
Pero repito, el el error NO sucede si ejecuto manualmente el script con un
correo como parámetro de CUALQUIER tamaño, sólo sucede si lo ejecuta
CommuniGate con correos GRANDES (de más de 100KB).
NPI de que estas intentando hacer, pero el error dice ni puedo escribir
en el standard output (normalmente consola), pregunto: que estas
intentando hacer y /adonde/ debiera redirigir su salida?
La respuesta a eso, te ayudara a solucionar tu problema..creo.
PD: Eso le sucede a /todos/ los niños que utilizan software no-libre y
de noche: se mean en la cama!
- --
Ricardo A.Frydman
Consultor en Tecnología Open Source - Administrador de Sistemas
jabber: [EMAIL PROTECTED] - http://www.eureka-linux.com.ar
SIP # 1-747-667-9534
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)
iD8DBQFECFLpkw12RhFuGy4RAppPAJ9/Lw206o3vgPTQLtBQ/XNAqlM/sgCfUMc1
G7yVDvHoJ7E+vzw8lGns6Qs=
=5v96
-END PGP SIGNATURE-
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
El Viernes, 3 de Marzo de 2006 07:52, Antonio Galicia escribió:
On 3/2/06, Iñaki [EMAIL PROTECTED] wrote:
Un problema similar sucede si uso awk en vez de sed:
-
--- cat $correo_original | awk
'ok{print}/^\s*$/{ok=1}' | formail -X $cabecera_original
-
---
Similar ... ¿tienes el texto del error?
Sí, usando esta expresión de sed:
Creo que no me di a entender. Si usas AWK, ¿qué mensaje de error aparece?
Perdona, me equivoqué, quería copiar lo de awk:
cat $correo_original | awk 'ok{print}/^\s*$/{ok=1}' | formail -X
$cabecera_original
11:00:02.67 1 ENQUEUERRULES [9570434] rule(myRule) action #1:
execute stderr: awk: cmd. line:1: (FILENAME=- FNR=139) fatal: print to
standard output failed (Broken pipe)\ncat: error de escritura: Tuber\237a
rota\n
Pero repito, el el error NO sucede si ejecuto manualmente el script con
un correo como parámetro de CUALQUIER tamaño, sólo sucede si lo ejecuta
CommuniGate con correos GRANDES (de más de 100KB).
En tal caso el problema no es con el script sino con quien lo invoca.
Sí, eso seguro. Mi pregunta de hecho no es sobre el script, sino sobre cuál
puede ser la causa de un error de tuberia rota con sed o awk para ver qué
relación puede tener con el servidor de correo que es quien invoca el script.
--
Saludos,
Toño
http://agc.com.mx/antoniogc/blog
--
y hasta aquí puedo leer...
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
El Viernes, 3 de Marzo de 2006 15:30, Ricardo Frydman Eureka! escribió:
Iñaki wrote:
El Jueves, 2 de Marzo de 2006 20:03, Antonio Galicia escribió:
On 3/2/06, Iñaki [EMAIL PROTECTED] wrote:
Un problema similar sucede si uso awk en vez de sed:
- --- cat $correo_original | awk
'ok{print}/^\s*$/{ok=1}' | formail -X $cabecera_original
- ---
Similar ... ¿tienes el texto del error?
Sí, usando esta expresión de sed:
cat $correo_original | sed '1,/^$/ d' | formail -X
$cabecera_original
el error sería éste:
-
--- 11:23:54.64 1 ENQUEUERRULES [9570439] rule(myRule)
action #1:
execute stderr: sed: couldn't write 72 items to {standard output}:
Broken pipe: error de escritura: Tuberia rota
-
-
Pero repito, el el error NO sucede si ejecuto manualmente el script con
un correo como parámetro de CUALQUIER tamaño, sólo sucede si lo ejecuta
CommuniGate con correos GRANDES (de más de 100KB).
NPI de que estas intentando hacer, pero el error dice ni puedo escribir
en el standard output (normalmente consola), pregunto: que estas
intentando hacer y /adonde/ debiera redirigir su salida?
Es un script que al que se le pasa como parámetro el la ruta a un determinado
fichero (un correo), lo lee y extrae sólo las cabeceras a otro fichero, así
que la escritura es a otro fichero. Y repito, desde consola funciona perfecto
y desde el servidor de correo sólo con ficheros pequeños.
PD: Eso le sucede a /todos/ los niños que utilizan software no-libre y
de noche: se mean en la cama!
Se lo con esas mismas palabras al departamento de informática de la empresa
dueña de dicho servidor. ;)
PD: Aprovecho para decir que en Linux no existe ningún servidor de correo como
solución final (SMTP, POP, IMAP, autenticación, reglas, interfaz web de
mantenimiento y configuración con Logs, gráficas y demás reportes, todo ello
integrado).
Que yo sepa sólo existe Kolab (www.kolab.org) que ofrece SMTP, POP, IMAP,
autenticación, agenda LDAP, groupware y un bonito interfaz web para
administrar usuarios y demás. Tiene muy buena pinta pero de momento no tiene
ni la mitad que el CommuniGate (he configurado ambos).
No me parece tan descabellado que empresas que apuesten por Linux tengan que
usar servidores de correo propietarios salvo que se pongan ellos mismos a
diseñarse su Postfix con LDAP, Cyrus y demás... y luego a administrarlo por
consola (estoy hablando de un caso de un servidor con más de 3000 usuarios, a
mí no me gustaría llevar su administración por consola). En el interfaz web
del CommuniGate pinchas en una pestañita y ves en tiempo real los correos que
se están enviando, recibiendo, gráficas, estadísticas y demás. Esas cosas se
agradecen y a veces son imprescindibles.
--
y hasta aquí puedo leer...
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Iñaki wrote:
El Viernes, 3 de Marzo de 2006 15:30, Ricardo Frydman Eureka! escribió:
Iñaki wrote:
El Jueves, 2 de Marzo de 2006 20:03, Antonio Galicia escribió:
On 3/2/06, Iñaki [EMAIL PROTECTED] wrote:
Un problema similar sucede si uso awk en vez de sed:
- --- cat $correo_original | awk
'ok{print}/^\s*$/{ok=1}' | formail -X $cabecera_original
- ---
Similar ... ¿tienes el texto del error?
Sí, usando esta expresión de sed:
cat $correo_original | sed '1,/^$/ d' | formail -X
$cabecera_original
el error sería éste:
-
--- 11:23:54.64 1 ENQUEUERRULES [9570439] rule(myRule)
action #1:
execute stderr: sed: couldn't write 72 items to {standard output}:
Broken pipe: error de escritura: Tuberia rota
-
-
Pero repito, el el error NO sucede si ejecuto manualmente el script con
un correo como parámetro de CUALQUIER tamaño, sólo sucede si lo ejecuta
CommuniGate con correos GRANDES (de más de 100KB).
NPI de que estas intentando hacer, pero el error dice ni puedo escribir
en el standard output (normalmente consola), pregunto: que estas
intentando hacer y /adonde/ debiera redirigir su salida?
Es un script que al que se le pasa como parámetro el la ruta a un determinado
fichero (un correo), lo lee y extrae sólo las cabeceras a otro fichero, así
que la escritura es a otro fichero. Y repito, desde consola funciona perfecto
y desde el servidor de correo sólo con ficheros pequeños.
PD: Eso le sucede a /todos/ los niños que utilizan software no-libre y
de noche: se mean en la cama!
Se lo con esas mismas palabras al departamento de informática de la empresa
dueña de dicho servidor. ;)
PD: Aprovecho para decir que en Linux no existe ningún servidor de correo
como
solución final (SMTP, POP, IMAP, autenticación, reglas, interfaz web de
mantenimiento y configuración con Logs, gráficas y demás reportes, todo ello
integrado).
Por supuesto! Y Dios permita que esto nunca llegue
Luego pediran una distribucion/escritorio totalmente integrado y asi
reinventaran las ventanas (no la rueda)...pero libres
Que yo sepa sólo existe Kolab (www.kolab.org) que ofrece SMTP, POP, IMAP,
autenticación, agenda LDAP, groupware y un bonito interfaz web para
administrar usuarios y demás. Tiene muy buena pinta pero de momento no tiene
ni la mitad que el CommuniGate (he configurado ambos).
No conozco CommuniGate, pero mis clientes quedan encantados con:
- - postfix
- - cyrus (solo IMAP)
- - usermin y webmin
- - horde (con imp, gollem y turba)
- - LDAP
(Generalmente, son desertores de las ventanitas /hartos/ de problemas)
No me parece tan descabellado que empresas que apuesten por Linux tengan que
Descabellado, no. Solo inutil.
usar servidores de correo propietarios salvo que se pongan ellos mismos a
diseñarse su Postfix con LDAP, Cyrus y demás...
Para que diseñar uno propio si ya existe uno?
y luego a administrarlo por
consola (estoy hablando de un caso de un servidor con más de 3000 usuarios, a
mí no me gustaría llevar su administración por consola).
Yo detestaria manejar eso por web ;)
En el interfaz web
del CommuniGate pinchas en una pestañita y ves en tiempo real los correos que
se están enviando, recibiendo, gráficas, estadísticas y demás.
Eso si que es util!
Esas cosas se
agradecen y a veces son imprescindibles.
Como decian en los tiempos de mi abuela: la juventud esta perdida.
- --
Ricardo A.Frydman
Consultor en Tecnología Open Source - Administrador de Sistemas
jabber: [EMAIL PROTECTED] - http://www.eureka-linux.com.ar
SIP # 1-747-667-9534
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)
iD8DBQFECF/Okw12RhFuGy4RAqzWAJ4/kvcrvA3tzZ4bjv2ljqD5cS7BSgCfdAVc
8tvqdXE5CQhxeis/Zcb623Y=
=ecas
-END PGP SIGNATURE-
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
On 3/3/06, Ricardo Frydman Eureka! [EMAIL PROTECTED] wrote: PD: Aprovecho para decir que en Linux no existe ningún servidor de correo como solución final (SMTP, POP, IMAP, autenticación, reglas, interfaz web de mantenimiento y configuración con Logs, gráficas y demás reportes, todo ello integrado). Por supuesto! Y Dios permita que esto nunca llegue ¿? ¿Razón? Luego pediran una distribucion/escritorio totalmente integrado y asi reinventaran las ventanas (no la rueda)...pero libres ¿No es lo que hace (casí) Ubuntú? No conozco CommuniGate, pero mis clientes quedan encantados con: - - postfix - - cyrus (solo IMAP) - - usermin y webmin - - horde (con imp, gollem y turba) - - LDAP (Generalmente, son desertores de las ventanitas /hartos/ de problemas) Perfecto, espero que tienes documentado (un HOWTO) de como se hace todo esto y puedes ponerlo al servicio de la comunidad como muestra de agradecimiento al trabajo en que basas el tuyo propio. Y lo vas a hacer público en los próximos dias, ¿cierto? usar servidores de correo propietarios salvo que se pongan ellos mismos a diseñarse su Postfix con LDAP, Cyrus y demás... Para que diseñar uno propio si ya existe uno? Y se llama ... ¿? Yo detestaria manejar eso por web ;) ¿Y qué fue del usermin y webmin? del CommuniGate pinchas en una pestañita y ves en tiempo real los correos que se están enviando, recibiendo, gráficas, estadísticas y demás. Eso si que es util! Ug... debe serlo para alguien y está dispuesto a pagar por tenerlo. ¿No ves una oportunidad de negocio? Si tienes armado todo el proceso que le ofreces a tus clientes se puede agregar este tipo de estadísticas y venderlo como un paquete completo. ¿No? Esas cosas se agradecen y a veces son imprescindibles. Como decian en los tiempos de mi abuela: la juventud esta perdida. No hagas caso, mi agüela decía lo mismo... y la abuela de ella otro tanto. -- Saludos, Toño http://agc.com.mx/antoniogc/blog
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
El Viernes, 3 de Marzo de 2006 16:25, Ricardo Frydman Eureka! escribió: Iñaki wrote: PD: Aprovecho para decir que en Linux no existe ningún servidor de correo como solución final (SMTP, POP, IMAP, autenticación, reglas, interfaz web de mantenimiento y configuración con Logs, gráficas y demás reportes, todo ello integrado). Por supuesto! Y Dios permita que esto nunca llegue Entonces Linux sólo será para administradores experimentados que se montan ellos mismo su SMTP, POP, IMAP y absolutamente todo, y que sólo dicho administrador sea capaz de meter mano a lo que ha creado. El empresario que tiene una oficina con 4 empleados y se atreve con cosas de Windows más gráficas que se olvide de Linux, ese no es de los buenos, que use soluciones integradas y finales de Microsoft o que pague a un consultor para que le diseñe un servidor de correo a medida. Yo no creo que haya ningún problema en que Linux como servidor se haga más fácil y accesible a gente que si no se acaba de acercar es porque teme a la consola. Que yo sepa sólo existe Kolab (www.kolab.org) que ofrece SMTP, POP, IMAP, autenticación, agenda LDAP, groupware y un bonito interfaz web para administrar usuarios y demás. Tiene muy buena pinta pero de momento no tiene ni la mitad que el CommuniGate (he configurado ambos). No conozco CommuniGate, pero mis clientes quedan encantados con: - postfix - cyrus (solo IMAP) - usermin y webmin - horde (con imp, gollem y turba) - LDAP Por supuesto, pero que quede bien claro que sólo tú serás capaz de meter mano a ese servidor que has diseñado a tu gusto, y que seguro que es estupendo, pero si algún día necesitan asistencia y tú no puedes atenderles tendrán que gastarse un dineral en contratar a un consultor para que investigue cómo funciona el invento. Por cierto, kolab.org es precisamente: - postfix - cyrus (POP e IMAP) - OpenLDAP - Interfaz web de administración - Clientes como horde o Kontact y conectores para Mozilla y propietarios para Outlook (para el tema de groupware). (Generalmente, son desertores de las ventanitas /hartos/ de problemas) En realidad yo sólo hablo de un interfaz web de administración centralizada, nada más. No me parece tan descabellado que empresas que apuesten por Linux tengan que Descabellado, no. Solo inutil. usar servidores de correo propietarios salvo que se pongan ellos mismos a diseñarse su Postfix con LDAP, Cyrus y demás... Para que diseñar uno propio si ya existe uno? ¿A cuál te refieres? y luego a administrarlo por consola (estoy hablando de un caso de un servidor con más de 3000 usuarios, a mí no me gustaría llevar su administración por consola). Yo detestaria manejar eso por web ;) Hombre, pero una web es accesible desde cualquier sitio, navegador y SO, y si te enseña cositas gráficamente de forma más agradable que una fría consola... ¿qué tiene de malo? En el interfaz web del CommuniGate pinchas en una pestañita y ves en tiempo real los correos que se están enviando, recibiendo, gráficas, estadísticas y demás. Eso si que es util! Y hasta divertido ;) Esas cosas se agradecen y a veces son imprescindibles. Como decian en los tiempos de mi abuela: la juventud esta perdida. Tranquilo, esa frase la diremos todos algún día ;) -- y hasta aquí puedo leer...
sed: couldn't write 72 items to {standard output}: Broken pipe
Hola.
Estoy desarrollando para una empresa unos pequeños scripts a medida para
complementar a un servidor de correo propietario (CommuniGate) y dotarle de
funcionalidades que requieren.
Básicamente existen unas Queue Rules es las que puedes definir una Action
que sea ejecutar un programa externo (mi script) y pasarle como parámetro el
path al fichero (correo) que se está procesando (el que ha cumplido la regla
y sufre la Action).
He hecho multitud de pruebas y el script funciona perfectamente ejecutado
manualmente por consola pasándole como parámetros el path a ficheros de
correo de todos los tamaños (ningún error).
Pero cuando está en producción y es invocado por el CommuniGate resulta que
falla al procesar archivos de correo de cierto tamaño (más de 100KB más o
menos) y los Logs del servidor de correo muestran un error tal que así:
11:23:54.64 1 ENQUEUERRULES [9570439] rule(myRule) action #1:
execute stderr: sed: couldn't write 72 items to {standard output}:
Broken pipet: error de escritura: Tuberia rota
El fragmento de mi script que usa sed es éste:
cat $correo_original | sed '1,/^$/ d' | formail -X $cabecera_original
Lo que hace es coger el correo original y extraer sólo las cabeceras
(eliminando el cuerpo y una cabecera especial que añade al inicio el propio
CommuniGate mientras el correo está encolado).
Nota: Agradezco de nuevo la ayuda a la gente que me indicó como hacer
precisamente esta parte del script.
Un problema similar sucede si uso awk en vez de sed:
cat $correo_original | awk 'ok{print}/^\s*$/{ok=1}' | formail -X
$cabecera_original
En cualquier caso, quiero dejar BIEN CLARO que NO es un problema del script,
ya que ejecutado en consola funciona PERFECTAMENTE con ficheros de 10MB si
hace falta.
Este error se produce SOLO cuando el script es invocado por el CommuniGate
porque un correo cumple la regla que lo llama, y dicho correo supera los
100KB (o similar). Si el correo es menor el script funciona perfectamente.
Yo me temo que tal vez el CommuniGate establezca un tiempo límite a los
programas externos a los que llama y que tal vez por eso los correos grandes
no tienen tiempo de ser procesados por mi script que es interrumpido y causa
el error. No lo sé. En cualquier caso hoy mismo he escrito al soporte de
dicho servidor de correo, a ver qué me dicen mañana.
No obstante, me gustaría saber qué significa el error de los Logs:
11:23:54.64 1 ENQUEUERRULES [9570439] rule(myRule) action #1:
execute stderr: sed: couldn't write 72 items to {standard output}:
Broken pipet: error de escritura: Tuberia rota
¿Cuándo se puede producir este error de sed?
¿por qué se podría producir llamado desde el CommuniGate y NUNCA si lo ejecuto
manualmente en consola?
¿tal vez porque es interrumpido?
¿o por un problema de algún buffer?
Desde ya gracias por cualquier ayuda y un saludo a todos.
PD: Por si pudiese ser de utilidad escribo las variables de entorno desde
consola como root y también las variables de entorno desde el CommuniGate
(las que hereda mi script cuando es ejecutado desde el CommuniGate). Yo no
veo nada raro pero quién sabe:
Variables de entorno en modo normal:
***
PWD=/root
[EMAIL PROTECTED]
HOSTNAME=mailbackup.dominio.com
LS_OPTIONS=--color=auto
QTDIR=/usr/lib/qt-2.3.0
LESSOPEN=|/usr/bin/lesspipe.sh %s
KDEDIR=/usr
USER=root
MACHTYPE=i386-redhat-linux-gnu
[EMAIL PROTECTED]
OLDPWD=/root
MAIL=/var/spool/mail/root
INPUTRC=/etc/inputrc
BASH_ENV=/root/.bashrc
[EMAIL PROTECTED]
GTK2_RC_FILES=/root/.gtkrc-2.0
LOGNAME=root
SHLVL=1
[EMAIL PROTECTED]
MM_CHARSET=ISO-8859-15
SHELL=/bin/bash
USERNAME=root
HOSTTYPE=i386
OSTYPE=linux-gnu
HISTSIZE=1000
TERM=xterm
HOME=/root
PATH=/usr/local/sbin:/usr/sbin:/sbin:/usr/local/bin:/usr/sbin:/sbin:/usr/bin:/bin:/usr/X11R6/bin/:/usr/games/:/root/bin:/root/bin
Variables de entorno desde CommuniGate:
***
PWD=/var/CommuniGate
LD_ASSUME_KERNEL=2.4.1
BOOT_FILE=/boot/vmlinuz-2.4.20
HOSTNAME=mailbackup.dominio.com
CONSOLE=/dev/console
PREVLEVEL=N
AUTOBOOT=YES
CONFIRM=
runlevel=3
MACHTYPE=i386-redhat-linux-gnu
[EMAIL PROTECTED]
SHLVL=3
previous=N
BOOT_IMAGE=linux-2.4.20
SHELL=/bin/bash
HOSTTYPE=i386
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
On 3/2/06, Iñaki [EMAIL PROTECTED] wrote:
Un problema similar sucede si uso awk en vez de sed:
cat $correo_original | awk 'ok{print}/^\s*$/{ok=1}' | formail -X
$cabecera_original
Similar ... ¿tienes el texto del error?
--
Saludos,
Toño
http://agc.com.mx/antoniogc/blog
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
El Jueves, 2 de Marzo de 2006 20:03, Antonio Galicia escribió:
On 3/2/06, Iñaki [EMAIL PROTECTED] wrote:
Un problema similar sucede si uso awk en vez de sed:
-
--- cat $correo_original | awk
'ok{print}/^\s*$/{ok=1}' | formail -X $cabecera_original
-
---
Similar ... ¿tienes el texto del error?
Sí, usando esta expresión de sed:
cat $correo_original | sed '1,/^$/ d' | formail -X $cabecera_original
el error sería éste:
11:23:54.64 1 ENQUEUERRULES [9570439] rule(myRule) action #1:
execute stderr: sed: couldn't write 72 items to {standard output}:
Broken pipe: error de escritura: Tuberia rota
--
Pero repito, el el error NO sucede si ejecuto manualmente el script con un
correo como parámetro de CUALQUIER tamaño, sólo sucede si lo ejecuta
CommuniGate con correos GRANDES (de más de 100KB).
--
y hasta aquí puedo leer...
Re: sed: couldn't write 72 items to {standard output}: Broken pipe
On 3/2/06, Iñaki [EMAIL PROTECTED] wrote:
Un problema similar sucede si uso awk en vez de sed:
-
--- cat $correo_original | awk
'ok{print}/^\s*$/{ok=1}' | formail -X $cabecera_original
-
---
Similar ... ¿tienes el texto del error?
Sí, usando esta expresión de sed:
Creo que no me di a entender. Si usas AWK, ¿qué mensaje de error aparece?
Pero repito, el el error NO sucede si ejecuto manualmente el script con un
correo como parámetro de CUALQUIER tamaño, sólo sucede si lo ejecuta
CommuniGate con correos GRANDES (de más de 100KB).
En tal caso el problema no es con el script sino con quien lo invoca.
--
Saludos,
Toño
http://agc.com.mx/antoniogc/blog
Re: install krita. Broken pipe. Do I need a plumber?
Kevin B. McCarty wrote: Hi Scott, Scott wrote: But seriously folks... What's a broken pipe? A pipe is this keyboard symbol: | That part I knew. :-) When you want to use the output of one command as the input of another command, you pipe it, like this: ls -l | sort And that part too.. :-) A broken pipe happens when the command reading input dies unexpectedly, for instance (supposing I was a very fast typist) if I meanwhile ran killall -9 sort in another xterm before ls was finished. Gotcha! Thanks dpkg: error processing /var/cache/apt/archives/krita_1%3a1.4.90.1-1_i386.deb (--unpack): trying to overwrite `/usr/share/mimelnk/image/x-raw.desktop', which is also in package digikam dpkg-deb: subprocess paste killed by signal ( *Broken pipe* ) Errors were encountered while processing: /var/cache/apt/archives/krita_1%3a1.4.90.1-1_i386.deb E: Sub-process /usr/bin/dpkg returned an error code (1) In this case, the program that dies is dpkg, and the reason it dies is stated on the second line I quoted above: the already-installed digikam package and the new krita package that you are trying to install contain the same file. This is not supposed to happen, and is therefore a bug in one of the two packages. Since your krita package came from experimental, it's probably the one at fault. Please file a bug against it. Done. :-) In the meantime, to work around the problem, you can run dpkg -i --force-overwrite /var/cache/apt/archives/krita_*.deb to tell dpkg that you want it to overwrite the common file instead of griping about it. Thanks! Now I don't have to confuse the plumber. ;-) -- Scott www.angrykeyboarder.com © 2006 angrykeyboarder™ Elmer Fudd. All Wights Wesewved
Re: install krita. Broken pipe. Do I need a plumber?
Hi Scott, Scott wrote: But seriously folks... What's a broken pipe? A pipe is this keyboard symbol: | When you want to use the output of one command as the input of another command, you pipe it, like this: ls -l | sort A broken pipe happens when the command reading input dies unexpectedly, for instance (supposing I was a very fast typist) if I meanwhile ran killall -9 sort in another xterm before ls was finished. dpkg: error processing /var/cache/apt/archives/krita_1%3a1.4.90.1-1_i386.deb (--unpack): trying to overwrite `/usr/share/mimelnk/image/x-raw.desktop', which is also in package digikam dpkg-deb: subprocess paste killed by signal ( *Broken pipe* ) Errors were encountered while processing: /var/cache/apt/archives/krita_1%3a1.4.90.1-1_i386.deb E: Sub-process /usr/bin/dpkg returned an error code (1) In this case, the program that dies is dpkg, and the reason it dies is stated on the second line I quoted above: the already-installed digikam package and the new krita package that you are trying to install contain the same file. This is not supposed to happen, and is therefore a bug in one of the two packages. Since your krita package came from experimental, it's probably the one at fault. Please file a bug against it. In the meantime, to work around the problem, you can run dpkg -i --force-overwrite /var/cache/apt/archives/krita_*.deb to tell dpkg that you want it to overwrite the common file instead of griping about it. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
dcc: write(MTA socket,88): Broken pipe
I have sarge running with the dcc-client and spamd running. I have been getting the following messages in my syslog - they appear after spamd completes processing the message: Dec 2 07:20:30 debian dccifd[5800]: write(MTA socket,88): Broken pipe Note that this does not appear after every message that spamd processes. Does this indicate a problem with dcc or spamd? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Apt-get: Broken Pipe
Hi! I tried to install an unofficial debian package (AMPortal) by adding a new ressource to resources.list. But the installation failed with the error message: --- apt-get install amportal asterisk-sounds-amportal [...] Fetched 2622kB in 1s (2533kB/s) (Reading database ... 75051 files and directories currently installed.) Unpacking asterisk-sounds-amportal (from .../asterisk-sounds-amportal_1.10.009-42_all.deb) ... dpkg: error processing /var/cache/apt/archives/asterisk-sounds-amportal_1.10.009-42_all.deb (--unpac k): trying to overwrite `/usr/share/asterisk/sounds/ext-or-zero.gsm', which is also in package asterisk -sounds-extra dpkg-deb: subprocess paste killed by signal (Broken pipe) Errors were encountered while processing: /var/cache/apt/archives/asterisk-sounds-amportal_1.10.009-42_all.deb E: Sub-process /usr/bin/dpkg returned an error code (1) --- and I can't install any other packages anymore, because apt-get reports unmet dependencies: --- The following packages have unmet dependencies: amportal: Depends: asterisk-sounds-amportal but it is not going to be installed asterisk-config-amportal: Depends: asterisk-sounds-amportal but it is not going to be installed E: Unmet dependencies. Try 'apt-get -f install' with no packages (or specify a solution). --- How can I restore the initial state (or proceed with the installation)? btw: I'm running debian sarge, but installed a lot of newer packages through apt pinning. Greetings, Marcus
Re: Apt-get: Broken Pipe
On Mon, Nov 28, 2005 at 05:32:09PM +0800, Marcus Deluigi (intern) wrote: trying to overwrite `/usr/share/asterisk/sounds/ext-or-zero.gsm', which is also in package asterisk ^^^ the package conflicts with `asterisk'. and I can't install any other packages anymore, because apt-get reports unmet dependencies: --- The following packages have unmet dependencies: amportal: Depends: asterisk-sounds-amportal but it is not going to be installed asterisk-config-amportal: Depends: asterisk-sounds-amportal but it is not going to be installed E: Unmet dependencies. Try 'apt-get -f install' with no packages (or specify a solution). --- How can I restore the initial state (or proceed with the installation)? did you try apt-get -f install? You may need to apt-get remove amportal. -- Jon Dowland http://jon.dowland.name/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
RE: Apt-get: Broken Pipe
(intern) wrote: trying to overwrite `/usr/share/asterisk/sounds/ext-or-zero.gsm', which is also in package asterisk ^^^ the package conflicts with `asterisk'. and I can't install any other packages anymore, because apt-get reports unmet dependencies: --- The following packages have unmet dependencies: amportal: Depends: asterisk-sounds-amportal but it is not going to be installed asterisk-config-amportal: Depends: asterisk-sounds-amportal but it is not going to be installed E: Unmet dependencies. Try 'apt-get -f install' with no packages (or specify a solution). --- How can I restore the initial state (or proceed with the installation)? did you try apt-get -f install? You may need to apt-get remove amportal. I tried both with no success: apt-get -f gives me the same error message and apt-get remove drives me into a deep depenency list where I get to asterisk, which I don't want to remove. I even tried to move the file 'ext-or'zero.gsm' away, but with no effect.
Re: Apt-get: Broken Pipe
On Mon, Nov 28, 2005 at 05:32:09PM +0800, Marcus Deluigi (intern) wrote: Hi! I tried to install an unofficial debian package (AMPortal) by adding a new ressource to resources.list. But the installation failed with the error message: [...] The problem is that your unofficial package contains a file in an official debian package. Perhaps your unofficial package was not designed for debian? If it was designed for debian, it should have a conflict with the official package or require a different version (making this a bug in the unofficial package). Probably the safest solution is to install all asterisk-related packages from the same repository (i.e., install the package asterisk from the unofficial repository). If that's not there, you can get by creating a diversion for the conflicting file (see dpkg-divert(8), you will want to use the --package switch). This will allow install, but if the two versions differ, the asterisk package may break. Of course, it's just as likely the packages will conflict in some other respect, since clearly they have not been tested together. But it's worth a try, especially if asterisk-sounds-amportal just provides shared audio files. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
xmms does not play m3u playlists (broken pipe)
Hi, I am having trouble since looong to make xmms play m3u playlist files. Here is the problem: If I click on an m3u link on a webpage in mozilla, xmms starts but gives an error: Couldn't connect to host host Example: I tried clicking on vybrations la passion du son [haut-debit] mp3 audio and got the message: Couldn't connect to host freestr1.ovh.net:8004 Then I download the m3u file (listen.m3u) from the website and tried starting xmms from command line and got this error (and the dialog box saying couldn't connect to host): #-- $ xmms listen.m3u Message: device: default ** WARNING **: ctrl_write_packet(): Failed to send data: Broken pipe #-- Next, I played the file with mplayer and it worked perfectly: #-- $ mplayer -playlist listen.m3u #-- So, looks like there is a problem with xmms. Can anybody help in this? I am running Debian Testing and 2.6.12 kernel, and $ dpkg -l *xmms* | grep ^ii ii xmms 1.2.10+cvs20050809-4 Versatile X audio player ii xmms-arts 0.7.1-4 aRts Output plugin for xmms ii xmms-jackasyn 0.1-2JACK Output plugin for xmms ii xmms-mad 0.8-1mp3 input plugin for xmms based on libmad ii xmms-mp4 2.0.0-sarge0.2 a mp4/aac audio player for xmms ii xmms-skins 0.6-1Skins for XMMS Thanks, -HS -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: xmms does not play m3u playlists (broken pipe)
H.S. wrote: Hi, I am having trouble since looong to make xmms play m3u playlist files. Here is the problem: If I click on an m3u link on a webpage in mozilla, xmms starts but gives an error: Couldn't connect to host host Oops, forgot to mention the URL of the website on which I tried the example below. It is: http://dir.xiph.org/index.php Example: I tried clicking on vybrations la passion du son [haut-debit] mp3 audio and got the message: Couldn't connect to host freestr1.ovh.net:8004 Then I download the m3u file (listen.m3u) from the website and tried starting xmms from command line and got this error (and the dialog box saying couldn't connect to host): #-- $ xmms listen.m3u Message: device: default ** WARNING **: ctrl_write_packet(): Failed to send data: Broken pipe #-- Next, I played the file with mplayer and it worked perfectly: #-- $ mplayer -playlist listen.m3u #-- So, looks like there is a problem with xmms. Can anybody help in this? I am running Debian Testing and 2.6.12 kernel, and $ dpkg -l *xmms* | grep ^ii ii xmms 1.2.10+cvs20050809-4 Versatile X audio player ii xmms-arts 0.7.1-4 aRts Output plugin for xmms ii xmms-jackasyn 0.1-2JACK Output plugin for xmms ii xmms-mad 0.8-1mp3 input plugin for xmms based on libmad ii xmms-mp4 2.0.0-sarge0.2 a mp4/aac audio player for xmms ii xmms-skins 0.6-1Skins for XMMS Thanks, -HS -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: xmms does not play m3u playlists (broken pipe)
On Wed, 09 Nov 2005 17:52:21 -0500 H.S. [EMAIL PROTECTED] wrote: H.S. wrote: Hi, I am having trouble since looong to make xmms play m3u playlist files. Here is the problem: If I click on an m3u link on a webpage in mozilla, xmms starts but gives an error: Couldn't connect to host host Oops, forgot to mention the URL of the website on which I tried the example below. It is: http://dir.xiph.org/index.php I just grabbed/listened to 2 sites, and they work. I just inserted your link into xmms, and I'm listening right now -- Rodney D. Myers [EMAIL PROTECTED] Registered Linux User #96112 ICQ#: AIM#: YAHOO: 18002350 mailman452 mailman42_5 They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety. Ben Franklin - 1759 signature.asc Description: PGP signature
Broken pipe when attempting to upgrade to Etch
With xorg now in etch, I attempt to upgrade from Sarge via apt-get dist-upgrade with preferences set to testing, and the process halts with error processing .../x11-common_6.8.2.dfsg.1-6_all.deb (unpack): trying to overwrite '/etc/X11/Xsession', which is also in package xorg-common process killed by signal (Broken Pipe) Any Ideas how to work around? -- Regards, Dave -- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: dpkg-deb: subprocess paste killed by signal (Broken pipe)
Am Donnerstag, 18. November 2004 00:36 schrieb Jochen von der Wall: Am Mittwoch, 17. November 2004 21:33 schrieb Michael Bienia: Gute Frage. Welche Ausgabe erhälst du bei dpkg -l 'libuuid1*' # dpkg -l 'libuuid1*' Desired=Unknown/Install/Remove/Purge/Hold Status=Not/Installed/Config-files/Unpacked/Failed-config/Half-install ed |/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad) ||/ Name VersionDescription +++ ii libuuid1 1.34+1.35-WIP- Universally unique id library ii libuuid1-udeb 1.35-1 Universally unique id library # Hi Michael, das war der richtige Hinweis: ii libuuid1-udeb 1.35-1 Universally unique id library. Ich habe in dselect dieses Paket einschl. evtl. Konf.-Dateien gelöscht und gut war's. Danke und schönes Wochenende. Grüße Jochen
Re: dpkg-deb: subprocess paste killed by signal (Broken pipe)
On 17.Nov 2004 - 15:52:03, Jochen von der Wall wrote: Hallo, ich bekomme beim apt-get upgrade von Sarge, Kernel 2.6.8.1 die folgende Meldung: Unpacking replacement libuuid1 ... dpkg: error processing /cdrom//pool/main/e/e2fsprogs/libuuid1_1.35-6_i386.deb (--unpack): trying to overwrite `/lib/libuuid.so.1.2', which is also in package libuuid1-udeb dpkg-deb: subprocess paste killed by signal (Broken pipe) Errors were encountered while processing: /cdrom//pool/main/e/e2fsprogs/libuuid1_1.35-6_i386.deb E: Sub-process /usr/bin/dpkg returned an error code (1) Was muß ich machen, daß es weitergeht? Bugreport schreiben ist das beste. Es gibt die Moeglichkeit mittels dpkgs --force-Option zu arbeiten, aber IMHO ist es noetig den Maintainer zu informieren, schliesslich ist das ein ziemlich grober Fehler. Wobei es mich wundert, dass er noch nicht bekannt ist... Andreas -- Rocky's Lemma of Innovation Prevention: Unless the results are known in advance, funding agencies will reject the proposal. -- Haeufig gestellte Fragen und Antworten (FAQ): http://www.de.debian.org/debian-user-german-FAQ/ Zum AUSTRAGEN schicken Sie eine Mail an [EMAIL PROTECTED] mit dem Subject unsubscribe. Probleme? Mail an [EMAIL PROTECTED] (engl)
Re: dpkg-deb: subprocess paste killed by signal (Broken pipe)
On 2004-11-17 17:02:01 +0100, Andreas Pakulat wrote: On 17.Nov 2004 - 15:52:03, Jochen von der Wall wrote: Hallo, ich bekomme beim apt-get upgrade von Sarge, Kernel 2.6.8.1 die folgende Meldung: Unpacking replacement libuuid1 ... dpkg: error processing /cdrom//pool/main/e/e2fsprogs/libuuid1_1.35-6_i386.deb (--unpack): trying to overwrite `/lib/libuuid.so.1.2', which is also in package libuuid1-udeb dpkg-deb: subprocess paste killed by signal (Broken pipe) Errors were encountered while processing: /cdrom//pool/main/e/e2fsprogs/libuuid1_1.35-6_i386.deb E: Sub-process /usr/bin/dpkg returned an error code (1) Was muß ich machen, daß es weitergeht? Bugreport schreiben ist das beste. Es gibt die Moeglichkeit mittels dpkgs --force-Option zu arbeiten, aber IMHO ist es noetig den Maintainer zu informieren, schliesslich ist das ein ziemlich grober Fehler. Wobei es mich wundert, dass er noch nicht bekannt ist... Was hier wurdert ist das udeb. Soviel ich weiß werden die nur vom Debian-Installer genutzt und haben auf einem normalen System nichts verloren. Michael -- Haeufig gestellte Fragen und Antworten (FAQ): http://www.de.debian.org/debian-user-german-FAQ/ Zum AUSTRAGEN schicken Sie eine Mail an [EMAIL PROTECTED] mit dem Subject unsubscribe. Probleme? Mail an [EMAIL PROTECTED] (engl)
Re: dpkg-deb: subprocess paste killed by signal (Broken pipe)
Am Mittwoch, 17. November 2004 18:31 schrieb Michael Bienia: Bugreport schreiben ist das beste. Es gibt die Moeglichkeit mittels dpkgs --force-Option zu arbeiten, aber IMHO ist es noetig den Maintainer zu informieren, schliesslich ist das ein ziemlich grober Fehler. Wobei es mich wundert, dass er noch nicht bekannt ist... Was hier wurdert ist das udeb. Soviel ich weiß werden die nur vom Debian-Installer genutzt und haben auf einem normalen System nichts verloren. Michael ... und was heißt das für mich? Die installierten Pakete sind vom Lehmanns-CD-Satz Frühjahr 2004, Linuxtag-DVD und downloads von http://http.us.debian.org testing/main Jo
Re: dpkg-deb: subprocess paste killed by signal (Broken pipe)
On 17.Nov 2004 - 18:31:42, Michael Bienia wrote: Was hier wurdert ist das udeb. Soviel ich weiß werden die nur vom Debian-Installer genutzt und haben auf einem normalen System nichts verloren. Ich kenne den Debian-Installer ja nun ueberhaupt nicht und dachte, dass der diese udeb's beim Installieren installiert (tolles Konstrukt) Andreas -- miracle: an extremely outstanding or unusual event, thing, or accomplishment. -- Webster's Dictionary -- Haeufig gestellte Fragen und Antworten (FAQ): http://www.de.debian.org/debian-user-german-FAQ/ Zum AUSTRAGEN schicken Sie eine Mail an [EMAIL PROTECTED] mit dem Subject unsubscribe. Probleme? Mail an [EMAIL PROTECTED] (engl)
Re: dpkg-deb: subprocess paste killed by signal (Broken pipe)
On 2004-11-17 19:31:18 +0100, Andreas Pakulat wrote: On 17.Nov 2004 - 18:31:42, Michael Bienia wrote: Was hier wurdert ist das udeb. Soviel ich weiß werden die nur vom Debian-Installer genutzt und haben auf einem normalen System nichts verloren. Ich kenne den Debian-Installer ja nun ueberhaupt nicht und dachte, dass der diese udeb's beim Installieren installiert (tolles Konstrukt) So genau kenne ich mich mit dem Installer auch nicht aus. Soweit ich das verstanden habe sind die udebs abgespeckte Pakete extra für den Installer für dessen eigenen Gebrauch. Deshalb gibt es auch nicht jedes Paket als udeb. Der Installer nutzt dann debootstrap um die normalen Pakete auf das Zielsystem zu installieren. Michael -- Haeufig gestellte Fragen und Antworten (FAQ): http://www.de.debian.org/debian-user-german-FAQ/ Zum AUSTRAGEN schicken Sie eine Mail an [EMAIL PROTECTED] mit dem Subject unsubscribe. Probleme? Mail an [EMAIL PROTECTED] (engl)
Re: dpkg-deb: subprocess paste killed by signal (Broken pipe)
On 2004-11-17 19:17:06 +0100, Jochen von der Wall wrote: Am Mittwoch, 17. November 2004 18:31 schrieb Michael Bienia: Bugreport schreiben ist das beste. Es gibt die Moeglichkeit mittels dpkgs --force-Option zu arbeiten, aber IMHO ist es noetig den Maintainer zu informieren, schliesslich ist das ein ziemlich grober Fehler. Wobei es mich wundert, dass er noch nicht bekannt ist... Was hier wurdert ist das udeb. Soviel ich weiß werden die nur vom Debian-Installer genutzt und haben auf einem normalen System nichts verloren. Michael ... und was heißt das für mich? Gute Frage. Welche Ausgabe erhälst du bei dpkg -l 'libuuid1*' Die installierten Pakete sind vom Lehmanns-CD-Satz Frühjahr 2004, Linuxtag-DVD und downloads von http://http.us.debian.org testing/main Eigentlich sollte ja nur die libuuid1 installiert sein. Ich habe mir die Fehlermeldung nochmal genauer angesehen, und da will dpkg das Paket libuuid1 mit einer neueren Version aktualisieren, scheitert aber an dem Konflikt mit der libuuid1-udeb. Eigentlich sollte die bemängelte Lib zum Paket libuuid1 gehören. Was ergibt denn dpkg -S /lib/libuuid.so.1.2 Michael -- Haeufig gestellte Fragen und Antworten (FAQ): http://www.de.debian.org/debian-user-german-FAQ/ Zum AUSTRAGEN schicken Sie eine Mail an [EMAIL PROTECTED] mit dem Subject unsubscribe. Probleme? Mail an [EMAIL PROTECTED] (engl)
Re: dpkg-deb: subprocess paste killed by signal (Broken pipe)
Am Mittwoch, 17. November 2004 21:33 schrieb Michael Bienia: Gute Frage. Welche Ausgabe erhälst du bei dpkg -l 'libuuid1*' # dpkg -l 'libuuid1*' Desired=Unknown/Install/Remove/Purge/Hold | Status=Not/Installed/Config-files/Unpacked/Failed-config/Half-installed |/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad) ||/ Name VersionDescription +++-==-==- ii libuuid1 1.34+1.35-WIP- Universally unique id library ii libuuid1-udeb 1.35-1 Universally unique id library # Die installierten Pakete sind vom Lehmanns-CD-Satz Frühjahr 2004, Linuxtag-DVD und downloads von http://http.us.debian.org testing/main Eigentlich sollte ja nur die libuuid1 installiert sein. Ich habe mir die Fehlermeldung nochmal genauer angesehen, und da will dpkg das Paket libuuid1 mit einer neueren Version aktualisieren, scheitert aber an dem Konflikt mit der libuuid1-udeb. Eigentlich sollte die bemängelte Lib zum Paket libuuid1 gehören. Was ergibt denn dpkg -S /lib/libuuid.so.1.2 # dpkg -S /lib/libuuid.so.1.2 libuuid1-udeb: /lib/libuuid.so.1.2 # Michael Jochen
ssh: Write failed: Broken pipe
Pessoal, Alguém já passou por isso ou sabe como resolver? Tenho uma chave DSA no meu diretório .ssh, com as permissões corretas (0600 para o id_dsa e 0644 para id_dsa.pub). Quando executo o ssh sem ter um ssh-agent no ar, resulta em: $ ssh [EMAIL PROTECTED] Permission denied, please try again. Permission denied, please try again. Permission denied (publickey,password,keyboard-interactive). $ Quando executo o ssh com um ssh-agent no ar (eval `ssh-agent`), $ ssh [EMAIL PROTECTED] ssh_askpass: exec(/usr/bin/ssh-askpass): No such file or directory Write failed: Broken pipe $ Porém quando adiciono a chave ao ssh-agent (ssh-add), a autenticação funciona direitinho. Isso começou a ocorrer depois que fiz atualizações (source.list apontando para testing e unstable) Minhas configurações: $ uname -a Linux nargothrond 2.6.5-1 #1 Fri Apr 30 02:39:49 BRT 2004 i686 GNU/Linux $ dpkg -l ssh libc6 libpam0g libssl0.9.6 libwrap0 zlib1g libpam-modules debconf adduser Desejado=U=Desconhecido/Instalar/Remover/aPagar/H=Manter | status=Não/Instalado/arquiv.-Config./U=Descomp./Falhou-config/H=semi-inst. |/ Erro?=(nenhum)/H=Mantido/precisa-Reinst./X=os dois problemas (status,Erro: maiúsculas=ruim) ||/ NomeVersão Descrição +++-===-==- ||/ Nome Versão Descrição +++-==-==- ii ssh3.8p1-3Secure rlogin/rsh/rcp replacement (OpenSSH) ii libc6 2.3.2.ds1-12 GNU C Library: Shared libraries and Timezone ii libpam0g 0.76-19Pluggable Authentication Modules library ii libssl0.9.60.9.6c-2.woody SSL shared libraries ii libwrap0 7.6.dbs-2 Wietse Venema's TCP wrappers library ii zlib1g 1.2.1-5compression library - runtime ii libpam-modules 0.76-19Pluggable Authentication Modules for PAM ii debconf1.4.24 Debian configuration management system ii adduser3.52 Add and remove users and groups Rebaixando o ssh para stable (3.4p1-1.woody.3) o comportamento continua o mesmo. Alguém tem alguma sugestão? -- Feng Sian redbrick1 at terra dot com dot br Linux user #160781
Re: ssh: Write failed: Broken pipe
apt-get install ssh-askpass pensei que este bug ja estivesse corrigido, mas vamos averiguar isto On Tue, 11 May 2004 00:52:30 -0300 Feng Sian [EMAIL PROTECTED] wrote: Pessoal, Alguém já passou por isso ou sabe como resolver? Tenho uma chave DSA no meu diretório .ssh, com as permissões corretas (0600 para o id_dsa e 0644 para id_dsa.pub). Quando executo o ssh sem ter um ssh-agent no ar, resulta em: $ ssh [EMAIL PROTECTED] Permission denied, please try again. Permission denied, please try again. Permission denied (publickey,password,keyboard-interactive). $ Quando executo o ssh com um ssh-agent no ar (eval `ssh-agent`), $ ssh [EMAIL PROTECTED] ssh_askpass: exec(/usr/bin/ssh-askpass): No such file or directory Write failed: Broken pipe $ Porém quando adiciono a chave ao ssh-agent (ssh-add), a autenticação funciona direitinho. Isso começou a ocorrer depois que fiz atualizações (source.list apontando para testing e unstable) Minhas configurações: $ uname -a Linux nargothrond 2.6.5-1 #1 Fri Apr 30 02:39:49 BRT 2004 i686 GNU/Linux $ dpkg -l ssh libc6 libpam0g libssl0.9.6 libwrap0 zlib1g libpam-modules debconf adduser Desejado=U=Desconhecido/Instalar/Remover/aPagar/H=Manter | status=Nã | o/Instalado/arquiv.-Config./U=Descomp./Falhou-config/H=semi-inst. |/ Erro?=(nenhum)/H=Mantido/precisa-Reinst./X=os dois problemas |(status,Erro: maiúsculas=ruim)|/ NomeVersão Descrição +++-===-==-== == ||/ Nome Versão Descrição +++-==-==-=== = ii ssh3.8p1-3Secure rlogin/rsh/rcp replacement (OpenSSH) ii libc6 2.3.2.ds1-12 GNU C Library: Shared libraries and Timezone ii libpam0g 0.76-19Pluggable Authentication Modules library ii libssl0.9.60.9.6c-2.woody SSL shared libraries ii libwrap0 7.6.dbs-2 Wietse Venema's TCP wrappers library ii zlib1g 1.2.1-5compression library - runtime ii libpam-modules 0.76-19Pluggable Authentication Modules for PAM ii debconf1.4.24 Debian configuration management system ii adduser3.52 Add and remove users and groups Rebaixando o ssh para stable (3.4p1-1.woody.3) o comportamento continua o mesmo. Alguém tem alguma sugestão? -- Feng Sian redbrick1 at terra dot com dot br Linux user #160781 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- --- Agney Lopes Roth Ferraz http://agney.linuxhome.com.br [EMAIL PROTECTED] Duvidas sobre debian: http://rautu.cipsga.org.br
Re: dpkg-deb: subprocesso paste morto por sinal (Broken pipe)
Em Qui, 2003-02-06 às 02:28, RC Lages escreveu: dpkg: erro processando /var/cache/apt/archives/kdebase-bin_4%3a3.1.0-0woody2_i386.deb (--unpack): tentando sobrescrever `/usr/bin/kdcop', que também está no pacote kdcop dpkg-deb: subprocesso paste morto por sinal (Broken pipe) Erros foram encontrados durante processamento de: /var/cache/apt/archives/kdebase-bin_4%3a3.1.0-0woody2_i386.deb E: Sub-process /usr/bin/dpkg returned an error code (1) debian:/home/lages# Alguém sabe como se resolve esse erro? Eu sei, nao sei se da melhor maneira: dpkg -i --force-all /var/cache/apt/archives/kdebase-bin_4%3a3.1.0-0woody2_i386.deb Isso ira instalar o pacote na mao, rasgando! Nao sou especialista, mas este tipo de erro apareceu muito quando passei a instalar pacotes nao-oficiais, e resolvi usando este parametro --force-all. Depois repeti o apt-get upgrade e o processo continuou. []'s -- ___ Esta mensagem foi certificada com livre de windows por Debian GNU/Linux http://www.debian-br.org signature.asc Description: PGP signature
dpkg-deb: subprocesso paste morto por sinal (Broken pipe)
debian:/home/lages# apt-get -u upgrade
Reading Package Lists... Done
Building Dependency Tree... Done
The following packages have been kept back
gnome gnome-terminal kaddressbook kdeprint korganizer ksysguard libnspr4 libnss3 mc
metacity mozilla mozilla-browser mozilla-mailnews
mozilla-psm
The following packages will be upgraded
kdebase-bin
1 packages upgraded, 0 newly installed, 0 to remove and 14 not upgraded.
Need to get 0B/485kB of archives. After unpacking 893kB will be used.
Do you want to continue? [Y/n]
(Lendo banco de dados ... 58861 arquivos e diretórios atualmente instalados.)
Preparando para substituir kdebase-bin 4:3.0.5a-0woody1 (usando
.../kdebase-bin_4%3a3.1.0-0woody2_i386.deb) ...
Descompactando substituto kdebase-bin ...
dpkg: erro processando /var/cache/apt/archives/kdebase-bin_4%3a3.1.0-0woody2_i386.deb
(--unpack):
tentando sobrescrever `/usr/bin/kdcop', que também está no pacote kdcop
dpkg-deb: subprocesso paste morto por sinal (Broken pipe)
Erros foram encontrados durante processamento de:
/var/cache/apt/archives/kdebase-bin_4%3a3.1.0-0woody2_i386.deb
E: Sub-process /usr/bin/dpkg returned an error code (1)
debian:/home/lages#
Alguém sabe como se resolve esse erro?
--
Obrigado pela especial atenção.
Paz, Vida Longa e Prospere!
_
| _ \ / ___| | |__ _ __ _ ___ ___ Ronaldo Cardozo Lages
| |_) | | | | / _` |/ _` |/ _ \/ __| E-mail [EMAIL PROTECTED]
| _ | |___ | |__| (_| | (_| | __/\__ \ ICQ #3640360 [licq 1.2.2]
|_| \_\\| |_\__,_|\__, |\___||___/ Porto Alegre, RS - Brasil
|___/
Debian-BR #407 [Woody] GNU/LinuxCounter #128171 GNU/LinuxMachine #178124
Projeto Software Livre RS [http://www.softwarelivre.rs.gov.br]
Re: [telnetd] ttloop: read: Broken pipe
Am Montag dem 22. Apr 2002 um 23:42 +0200 Uhr, meinte Hannes Lau: Hallo, Neustart dies: Die telnet-verbindung zum Rechner ist aeusserst lahm, aber nur beim Verbindungsaufbau... Also ich beobachte das derzeit. Nachdem ich den Rechner ueber Nacht mit Nichtbeachtung strafte und einem Neustart geht´s mal gut, mal weniger. Komisch... CU H.Lau -- - Hannes Lau| Mail: [EMAIL PROTECTED] Dierkower Höhe 31 | http://home.debitel.net/user/hannes.lau/default.html 18146 Rostock | Home- Office- Tel: +49 (0381) 69 65 18 - msg06490/pgp0.pgp Description: PGP signature
[telnetd] ttloop: read: Broken pipe
Hallo, Auf einem meiner Beistell-Rechner habe ich ein altes potato 2.2r2 auf woody 3.0 dist-upgraded. Soweit, sogut. Nun passiert nach dem Neustart dies: Die telnet-verbindung zum Rechner ist aeusserst lahm, aber nur beim Verbindungsaufbau (alles andere wie ssh, ftp geht wie bisher schnell/gut). Aus den log´s kommt dies: --- Apr 22 22:40:15 penti100 in.telnetd[576]: connect from penti600.lau.debitel.net Apr 22 22:40:35 penti100 telnetd[576]: ttloop: read: Broken pipe Apr 22 22:40:40 penti100 in.telnetd[577]: connect from penti333.hannes.linux --- Wo muss ich da ansetzen, um das Problem zu loesen? Was ist den ttloop? Danke fuer Antwort im voraus. CU H.Lau -- - Hannes Lau| Mail: [EMAIL PROTECTED] Dierkower Höhe 31 | http://home.debitel.net/user/hannes.lau/default.html 18146 Rostock | Home- Office- Tel: +49 (0381) 69 65 18 - msg06441/pgp0.pgp Description: PGP signature
potato to sid upgrade error 32 broken pipe
Had an extra hard drive lying around, decided to give sid a twirl. Installed minimal potato without X. Changed sources to point to unstable. Typed apt-get update; apt-get dist-upgrade as usual. Halfway thru upgrade I get: --- 90 packages upgraded, 17 newly installed, 0 toremove and 3 not upgraded. 3 packages not fully installed or removed. Need to get 0B/19.5MB of archives. After unpacking 13.0MB will be used. Do you want to continue? [Y/n] perl: error while loading shared libraries: libdb.so.3: cannot open shared object file: No such file or directory E: Write error - write (32 Broken pipe) E: Failure running sript /usr/sbin/dpkg-preconfigure --apt sid:~# Nevere ran into an error of this type. Should have I upgraded to woody first, then unstable? Any suggestions appreciated.
Re: potato to sid upgrade error 32 broken pipe
--- 90 packages upgraded, 17 newly installed, 0 toremove and 3 not upgraded. 3 packages not fully installed or removed. Need to get 0B/19.5MB of archives. After unpacking 13.0MB will be used. Do you want to continue? [Y/n] perl: error while loading shared libraries: libdb.so.3: cannot open shared object file: No such file or directory E: Write error - write (32 Broken pipe) E: Failure running sript /usr/sbin/dpkg-preconfigure --apt sid:~# It is not finding some perl modules that it needs. The required packages are probably in your cache. I suggest trying to manually install perl packages till you get one to configure (dpkg -i someperlpack.deb). Then get the others. Once you have a complete perl installation, try apt again. Mike
Re: potato to sid upgrade error 32 broken pipe
--- 90 packages upgraded, 17 newly installed, 0 toremove and 3 not upgraded. 3 packages not fully installed or removed. Need to get 0B/19.5MB of archives. After unpacking 13.0MB will be used. Do you want to continue? [Y/n] perl: error while loading shared libraries: libdb.so.3: cannot open shared object file: No such file or directory E: Write error - write (32 Broken pipe) E: Failure running sript /usr/sbin/dpkg-preconfigure --apt sid:~# Nevere ran into an error of this type. Should have I upgraded to woody first, then unstable? Any suggestions appreciated. Hello, I had these problems with perl when upgrading to woody. Just run apt-get dist-upgrade and apt-get -f install until both commands have finished (I had to do this over 20 times before my upgrade was complete!). Greetz, Sebastiaan
Re: potato to sid upgrade error 32 broken pipe
I had the same problem while doing exactly the same thing last week. Unfortunately, cant remember exactly how it was solved. There is a package in /var/cache/apt/archives that needs to installed manually using dpkg. I think it was a libc package or a db2 something. you can just do dpkg -i filename.deb sorry about the vagueness of the package names - it was 2am. Hope this helps Shri --- Sebastiaan [EMAIL PROTECTED] wrote: --- 90 packages upgraded, 17 newly installed, 0 toremove and 3 not upgraded. 3 packages not fully installed or removed. Need to get 0B/19.5MB of archives. After unpacking 13.0MB will be used. Do you want to continue? [Y/n] perl: error while loading shared libraries: libdb.so.3: cannot open shared object file: No such file or directory E: Write error - write (32 Broken pipe) E: Failure running sript /usr/sbin/dpkg-preconfigure --apt sid:~# Nevere ran into an error of this type. Should have I upgraded to woody first, then unstable? Any suggestions appreciated. Hello, I had these problems with perl when upgrading to woody. Just run apt-get dist-upgrade and apt-get -f install until both commands have finished (I had to do this over 20 times before my upgrade was complete!). Greetz, Sebastiaan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] __ Do You Yahoo!? Make international calls for as low as $.04/minute with Yahoo! Messenger http://phonecard.yahoo.com/
Re: potato to sid upgrade error 32 broken pipe
This is a problem alright. You may want to get debian-testing in on this, as a smooth potato - woody transition is getting more vital by the day. When woody freezes, there MUST be a smooth upgrade path from potato On Wed, 1 Aug 2001, Sebastiaan wrote: --- 90 packages upgraded, 17 newly installed, 0 toremove and 3 not upgraded. 3 packages not fully installed or removed. Need to get 0B/19.5MB of archives. After unpacking 13.0MB will be used. Do you want to continue? [Y/n] perl: error while loading shared libraries: libdb.so.3: cannot open shared object file: No such file or directory E: Write error - write (32 Broken pipe) E: Failure running sript /usr/sbin/dpkg-preconfigure --apt sid:~# Nevere ran into an error of this type. Should have I upgraded to woody first, then unstable? Any suggestions appreciated. Hello, I had these problems with perl when upgrading to woody. Just run apt-get dist-upgrade and apt-get -f install until both commands have finished (I had to do this over 20 times before my upgrade was complete!). Greetz, Sebastiaan -- Pardon me, but you have obviously mistaken me for someone who gives a damn. email [EMAIL PROTECTED]
broken pipe
In attempting to dpkg -i libgtk1.2_1.2.6-1.deb, the current potato libgtk, I'm getting the following: dpkg-deb: subprocess paste killed by signal (Broken pipe) dpkg: error processing libgtk1.2_1.2.6-1.deb) ... subprocess dpkg-deb --fsys-tarfile returned error exit status 2 Could someone please clue me in to as to why this is happening? When the file was originally downloaded it stopped 6 bytes short, which I in turn did a 'reget' for and it is now complete as far as I can tell. Thanks in advance.. -cheshire
Re: broken pipe
--- [EMAIL PROTECTED] wrote: In attempting to dpkg -i libgtk1.2_1.2.6-1.deb, the current potato libgtk, I'm getting the following: dpkg-deb: subprocess paste killed by signal (Broken pipe) dpkg: error processing libgtk1.2_1.2.6-1.deb) ... subprocess dpkg-deb --fsys-tarfile returned error exit status 2 Could someone please clue me in to as to why this is happening? When the file was originally downloaded it stopped 6 bytes short, which I in turn did a 'reget' for and it is now complete as far as I can tell. Thanks in advance.. Looks like the file is broken nonetheless. Try re-downloading it. Why don't you use apt-get? It makes like so much easier. Michel = Software is like sex; it's better when it's free -- Linus Torvalds If you continue running Windows, your system may become unstable. -- Windows 95 BSOD __ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com
Re: popularity-contest broken pipe
I should also note that I can run it from the command line no problem, it is just as a daily cron job that it has problems... Matt On Oct 21, 1999 at 05:41:26PM, Matthew W. Roberts wrote: I installed popularity-contest (stable) on my machine and every day I get the following message from cron: /etc/cron.daily/popularity-contest: /usr/bin/dpkg-awk: line 31: 2988 Broken pipe mawk $incf -f +/usr/lib/awk/dpkg-awk.lib -- --exebase=`basename $0` --exedir=`dirname $0` +$@
popularity-contest broken pipe
I installed popularity-contest (stable) on my machine and every day I get the following message from cron: /etc/cron.daily/popularity-contest: /usr/bin/dpkg-awk: line 31: 2988 Broken pipe mawk $incf -f +/usr/lib/awk/dpkg-awk.lib -- --exebase=`basename $0` --exedir=`dirname $0` +$@ Because of this, the report going out is blank. Any ideas why this might be happening? Thanks, -- Matthew Roberts Structural Engineering [EMAIL PROTECTED] Texas AM University What's another word for thesaurus? -- Steven Wright
Re: Attn: Newbies Re: Random Broken Pipe
On Sun, 22 Mar 1998 [EMAIL PROTECTED] wrote: ** Reply to note from Art Lemasters [EMAIL PROTECTED] Mon, 23 Mar 1998 00:20:29 -0600 (CST) default, or my bo was broken. In any case, less was an excellent idea for a default for manpage paging. Quite honestly I can't think of any DOS, Windows or OS/2 text file viewer program I've used over the last 10 years that didn't have the simple features of search and page up/down and backscroll. I couldn't believe my senses weren't playing tricks on me the first time I installed Debian Linux and found out the default file viewer didn't include page up/down! grin I could be wrong, but I think DOS's more command has no backscroll. Michael Beattie ([EMAIL PROTECTED]) --- Don't force it, use a bigger hammer. --- Debian GNU/Linux Ooohh You are missing out! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Random Broken Pipe
I get this: [79 jacinta:~] while true ; do echo -n . ; /usr/sbin/update-alternatives --display WindowMaker | grep -q 'WindowMaker-newstyle' ; done .Broken pipe ..Broken pipe ..Broken pipe ...Broken pipe ..^C do you see a problem with the statement? (It's just for test purposes, I'm only interested on the 'update-alternatives ... -newstyle' part) It says broken pipe randomly. Thanks Marcelo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Random Broken Pipe
Broken pipe means that the '| did not recieve any data. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Attn: Newbies Re: Random Broken Pipe
For those of you who are newbies like myself, I have found that using shell (e.g., bash, sh, etc.) manpages as a reference could probably save us a lot of time (and embarrassment) in the future--that is, if you, like myself, want to explore versatility through playing with configurations via the use of scripts. Some wonderful developer(s) fixed our manpage readers of the coming release (hamm) so that they will also scroll upwards to the beginning of any file we read. The manpages (accessed by typing man nameofcommand--without the quotes) can also be searched with the / foreslash command in the hamm version. Excellent work... You might also find an even easier beginners' tutorial for your shell (bash, sh, etc.) somewhere on the Web. Now, to _begin_ to practice what I just preached... ;-) The developers are doing some really good work! Soon, Debian LINUX will be everywhere. Art Shaleh enlightened: Broken pipe means that the '| did not recieve any data. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
On Sun, 22 Mar 1998, Art Lemasters wrote: Some wonderful developer(s) fixed our manpage readers of the coming release (hamm) so that they will also scroll upwards to the beginning of any file we read. The manpages (accessed by typing man nameofcommand--without the quotes) can also be searched with the / foreslash command in the hamm version. Excellent work... This has nothing to do with man. With bo, the pager that man used was more, and that couldn't go up. Now, the default is less, which does as you describe. This could have been changed manually at any time. You didn't have to wait for hamm. This might not have been done in hamm's man, I am not sure. Adam -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
Adam wrote the enlightenment: On Sun, 22 Mar 1998, Art Lemasters wrote: Some wonderful developer(s) fixed our manpage readers of the coming release (hamm) so that they will also scroll upwards to the beginning of any file we read. The manpages (accessed by typing man nameofcommand--without the quotes) can also be searched with the / foreslash command in the hamm version. Excellent work... This has nothing to do with man. With bo, the pager that man used was more, and that couldn't go up. Now, the default is less, which does as you describe. This could have been changed manually at any time. You didn't have to wait for hamm. This might not have been done in hamm's man, I am not sure. It worked that way for me in hamm. In bo, it was either not the default, or my bo was broken. In any case, less was an excellent idea for a default for manpage paging. Literature students tend to be more linear thinkers, whereas technology students tend to do more back- checking, so it was an excellent idea for a default whether implemented recently or long ago. I did not know that it could be used in bo at all. Thank you for the tip, Adam. Learning is the most important part of all of this for me, because I hope to lend some of that writer's linear thinking to developing documentation in the near future. Thanks again, Adam. And thanks, Shaleh, for your correction on this, also. Correct me to the list anytime. When it comes to incoming instruction, I try to be like a sponge while having the sensitivity of a lizard. ;-) Art Adam -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
Addressed to: Art Lemasters [EMAIL PROTECTED] Debian debian-user@lists.debian.org ** Reply to note from Art Lemasters [EMAIL PROTECTED] Sun, 22 Mar 1998 23:41:26 -0600 (CST) Some wonderful developer(s) fixed our manpage readers of the coming release (hamm) so that they will also scroll upwards to the beginning of any file we read. The manpages (accessed by typing man nameofcommand--without the quotes) can also be searched with the / foreslash command in the hamm version. Excellent work... You could also use: man program name | less What *really* puzzles me as a newcomer to the linux (debian) arena is why using less (or anything else that could backscroll and do searches) hasn't been made the ruddy default right from the begining. I have shaken my head in bemusement and frustration over this and the archaic wordstar text processors that seem to be included as default with more than one application I have here on my Debian CD. (one newbie user who is looking forward to a new stable debian release that hopely won't include such obvious archiac software) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
** Reply to note from Art Lemasters [EMAIL PROTECTED] Mon, 23 Mar 1998 00:20:29 -0600 (CST) default, or my bo was broken. In any case, less was an excellent idea for a default for manpage paging. Quite honestly I can't think of any DOS, Windows or OS/2 text file viewer program I've used over the last 10 years that didn't have the simple features of search and page up/down and backscroll. I couldn't believe my senses weren't playing tricks on me the first time I installed Debian Linux and found out the default file viewer didn't include page up/down! grin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
On Sun, 22 Mar 1998, Art Lemasters wrote: Some wonderful developer(s) fixed our manpage readers of the coming release (hamm) so that they will also scroll upwards to the beginning of any file we read. The manpages (accessed by typing man nameofcommand--without the quotes) can also be searched with the / foreslash command in the hamm version. Excellent work... This has nothing to do with man. With bo, the pager that man used was more, and that couldn't go up. Now, the default is less, which does as you describe. This could have been changed manually at any time. You didn't have to wait for hamm. tension rises as drum roll sounds fill the stadium ... Bang! (says big chinese cymbal) Excellent comment, now let's go all the way and tell them newbie people how to _do_ it :) in bask/ksh: export PAGER less in csh/tcsh: setenv PAGER less This could change the behaviour of more programs that use a pager (can't think of any now, except a few I wrote myself). cheers, Eric Meijer -- E.L. Meijer ([EMAIL PROTECTED]) | tel. office +31 40 2472189 Eindhoven Univ. of Technology | tel. lab. +31 40 2475032 Lab. for Catalysis and Inorg. Chem. (TAK) | tel. fax+31 40 2455054 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
On Mon, Mar 23, 1998 at 10:37:39AM +0100, E.L. Meijer (Eric) wrote: in bask/ksh: export PAGER less You want export PAGER=less I do believe. hamish (a tcsh user) -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd. Developers of music education software including Auralia Musition. 31 Elmhurst Road, Blackburn, Victoria Australia, 3130 Phone: +61 3 9894 4788 Fax: +61 3 9894 3362 USA Toll Free: 1-888-667-7839 Internet: http://www.rising.com.au/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
hamish wrote: On Mon, Mar 23, 1998 at 10:37:39AM +0100, E.L. Meijer (Eric) wrote: in bask/ksh: export PAGER less You want export PAGER=less I do believe. Yep, moreover, I meant `bash' and not `bask'. Maybe Monday mornings aren't the right time for emailing to the debian-user list for me %) Eric -- E.L. Meijer ([EMAIL PROTECTED]) | tel. office +31 40 2472189 Eindhoven Univ. of Technology | tel. lab. +31 40 2475032 Lab. for Catalysis and Inorg. Chem. (TAK) | tel. fax+31 40 2455054 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Random Broken Pipe
Marcelo E. Magallón [EMAIL PROTECTED] writes: [79 jacinta:~] while true ; do echo -n . ; /usr/sbin/update-alternatives --display WindowMaker | grep -q 'WindowMaker-newstyle' ; done .Broken pipe From the grep man page: -q, --quiet Quiet; suppress normal output. The scanning will stop on the first match. If grep exits (because it found a match) before update-alternatives is finished writing to the pipe, then update-alternatives will receive a SIGPIPE signal, and the shell will print Broken pipe after update-alternatives crashes. You can see this easily with something like `cat /dev/zero | true'. (true doesn't read any input before it exits.) do you see a problem with the statement? This is how it's supposed to work. Can someone else suggest a way to suppress the messages? -- Carey Evans http://home.clear.net.nz/pages/c.evans/ GNU GPL: The Source will be with you... always. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
I think that it is mostly historical (just about every Unix system since about 1970 has had 'more'). There are some other reasons that are now at least mostly not applicable. For example, 'more' will work on a printing terminal but I believe that 'less' will not. There might be other VDTs that can not use 'less'. Another 'problem' that surely would be rare anymore is that 'less' will use a great deal more memory than 'more' for large files. Take a look at 'zless' too. This nifty pager displays gzipped files too. -- best, -bill [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] from a 1996 Micro$loth ad campaign: The less you know about computers the more you want Micro$oft! See! They do get some things right! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
On Sun, 22 Mar 1998 [EMAIL PROTECTED] wrote: Some wonderful developer(s) fixed our manpage readers of the coming release (hamm) so that they will also scroll upwards to the beginning of any file we read. The manpages (accessed by typing man nameofcommand--without the quotes) can also be searched with the / foreslash command in the hamm version. Excellent work... You could also use: man program name | less What *really* puzzles me as a newcomer to the linux (debian) arena is why using less (or anything else that could backscroll and do searches) hasn't been made the ruddy default right from the begining. I have shaken my head in bemusement and frustration over this and the archaic wordstar text processors that seem to be included as default with more than one application I have here on my Debian CD. (one newbie user who is looking forward to a new stable debian release that hopely won't include such obvious archiac software) Well perhaps the reason is that a simple pager always works. The trouble with less is that it doesn't always work, particularly if you're struggling with keyboard and terminal settings. In particular, less has problems if the actual screen-size doesn't match up with what it thinks the screen-size is, particularly if it's repainting by clearing (which can fail) rather than scrolling dumbly. Cheers, -- David Wright, Open University, Earth Science Department, Milton Keynes MK7 6AA U.K. email: [EMAIL PROTECTED] tel: +44 1908 653 739 fax: +44 1908 655 151 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
On Sun, 22 Mar 1998 [EMAIL PROTECTED] wrote: [ snip ] : (one newbie user who is looking forward to a new stable debian release that : hopely won't include such obvious archiac software) Hmm ... well, I hope it includes new _and_ archaic software ... some of that stuff's quite useful! ;) (awk,sed,tr,cut,grep,sort,uniq,...) etc. -- Nathan Norman MidcoNet - 410 South Phillips Avenue - Sioux Falls, SD 57104 mailto://[EMAIL PROTECTED] http://www.midco.net finger [EMAIL PROTECTED] for PGP Key: (0xA33B86E9) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
On Mon, 23 Mar 1998, E.L. Meijer (Eric) wrote:
[ snipped discussion of less vs. more ]
: Excellent comment, now let's go all the way and tell them newbie people
: how to _do_ it :)
:
: in bask/ksh: export PAGER less
: in csh/tcsh: setenv PAGER less
Excellent support :) BTW, these lines could be placed in your
~/{.bash_profile,.profile,.cshrc} ... or /etc/{profile,cshrc}, although
then you're mucking about with etc config files :)
: This could change the behaviour of more programs that use a pager (can't
: think of any now, except a few I wrote myself).
One thing it will change is the behavior of 'make config' when
recompiling the kernel ... namely when you choose ``?'' to get more
information about that particular option. ``less'' will make you choose
``q'' to get back to the config menu. This is logical, but nevertheless
unexpected if you've made a kernel with ``more'' as the pager ...
FWIW, I always change my systems to use ``less'' as the default pager.
The users really seem to like it.
--
Nathan Norman
MidcoNet - 410 South Phillips Avenue - Sioux Falls, SD 57104
mailto://[EMAIL PROTECTED] http://www.midco.net
finger [EMAIL PROTECTED] for PGP Key: (0xA33B86E9)
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
On Mon, 23 Mar 1998, Bill Leach wrote: : I think that it is mostly historical (just about every Unix system since : about 1970 has had 'more'). : : There are some other reasons that are now at least mostly not : applicable. For example, 'more' will work on a printing terminal but I : believe that 'less' will not. There might be other VDTs that can not : use 'less'. Another 'problem' that surely would be rare anymore is that : 'less' will use a great deal more memory than 'more' for large files. : : Take a look at 'zless' too. This nifty pager displays gzipped files : too. Better yet, use ``lesspipe''. It's a shell script included in the latest ``less'' package ... you set it up like so ... (for bash/ksh anyway) eval $(lesspipe) Put this line in your .profile or /etc/profile. Now you can use ``less'' on anything ... There is no manpage for lesspipe, but it's mentioned in the `less'' manpage. -- Nathan Norman MidcoNet - 410 South Phillips Avenue - Sioux Falls, SD 57104 mailto://[EMAIL PROTECTED] http://www.midco.net finger [EMAIL PROTECTED] for PGP Key: (0xA33B86E9) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
On Mon, 23 Mar 1998 09:27:21 -0600 (CST) [EMAIL PROTECTED] wrote: One thing it will change is the behavior of 'make config' when recompiling the kernel ... namely when you choose ``?'' to get more information about that particular option. ``less'' will make you choose ``q'' to get back to the config menu. This is logical, but nevertheless unexpected if you've made a kernel with ``more'' as the pager ... I set LESS=-Miqe in my /etc/profile. (It could be in .bash_profile as well.) The -q Causes less to automatically exit the second time it reaches end-of-file. This makes the behavior mentioned above more like more. -- _ |_) _ |_ Robert D. Hilliard[EMAIL PROTECTED] |_) (_) |_) Palm City, FL USAPGP Key ID: A8E40EB9 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
On Mon, 23 Mar 1998, Bob Hilliard wrote: : On Mon, 23 Mar 1998 09:27:21 -0600 (CST) [EMAIL PROTECTED] wrote: : One thing it will change is the behavior of 'make config' when : recompiling the kernel ... namely when you choose ``?'' to get more : information about that particular option. ``less'' will make you choose : ``q'' to get back to the config menu. This is logical, but nevertheless : unexpected if you've made a kernel with ``more'' as the pager ... : : I set LESS=-Miqe in my /etc/profile. (It could be in : .bash_profile as well.) The -q Causes less to automatically exit : the second time it reaches end-of-file. This makes the behavior : mentioned above more like more. Well I'll be damned :) I learn something every day!! Thanks for the tip ... -- Nathan Norman MidcoNet - 410 South Phillips Avenue - Sioux Falls, SD 57104 mailto://[EMAIL PROTECTED] http://www.midco.net finger [EMAIL PROTECTED] for PGP Key: (0xA33B86E9) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
On Mon, Mar 23, 1998 at 09:38:53AM -0600, [EMAIL PROTECTED] wrote: On Mon, 23 Mar 1998, Bill Leach wrote: : I think that it is mostly historical (just about every Unix system since : about 1970 has had 'more'). : : There are some other reasons that are now at least mostly not : applicable. For example, 'more' will work on a printing terminal but I : believe that 'less' will not. There might be other VDTs that can not : use 'less'. Another 'problem' that surely would be rare anymore is that : 'less' will use a great deal more memory than 'more' for large files. : : Take a look at 'zless' too. This nifty pager displays gzipped files : too. Better yet, use ``lesspipe''. It's a shell script included in the latest ``less'' package ... you set it up like so ... (for bash/ksh anyway) eval $(lesspipe) Put this line in your .profile or /etc/profile. Now you can use ``less'' on anything ... There is no manpage for lesspipe, but it's mentioned in the `less'' manpage. Try man lesskey. Here's my .lesskey file: # $Header: /home/lee/RCS/.lesskey,v 1.1 1998/03/15 23:01:59 lee Exp $ #env LESS = -is LESSOPEN = |~/.lesspipe.sh %s Then you don't need the eval, and ANY version of less you use will read the configuration file. You already have a lesspipe script, so I didn't include mine. -- Lee Bradshaw [EMAIL PROTECTED] (preferred) Alantro Communications [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
DOOM - broken pipe
Has anyone gotten DOOM to work? Here, the program starts, but when I try to move the cursor, or press any ke, after the Menu comes up the program stops with the message 'broken pipe'. Any ideas? Orn Einar Hansen [EMAIL PROTECTED] [EMAIL PROTECTED] voice+fax; +46 035 217194 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: DOOM - broken pipe
Orn E. Hansen wrote: Has anyone gotten DOOM to work? Here, the program starts, but when I try to move the cursor, or press any ke, after the Menu comes up the program stops with the message 'broken pipe'. Any ideas? What version of doom are you using? Do you have doom-musserver installed, and if so, what version? Do you have a sound card? -- see shy jo -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: DOOM - broken pipe
Þann 01-Jan-98 skrifar Joey Hess: What version of doom are you using? Do you have doom-musserver installed, and if so, what version? Do you have a sound card? The problem was, that I am using NAS to network'ize my soundcard, and that meant doom couldn't acquire the card. Whenever I move the cursor in the menu, there is a sound, which is why it broke. There is a sndserver for NAS, and I compiled it. It seems to work ok, but it would be better if it was possible to set a variable to point to it, so you don't have to overwrite the original sndserver, which provides far better sound. Orn Einar Hansen [EMAIL PROTECTED] [EMAIL PROTECTED] voice+fax; +46 035 217194 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
