http://bugzilla.kernel.org/show_bug.cgi?id=6402
Don't refer to external links with bug description. This make searching for duplicates completely useless...
I think that its bug in kernel. I add this bug to your bugtrack for information. Ok. I may copy info here =) After start download file its stop after 4-10 mb and network is down. Interface ppp0 is UP. No new dmeseg or debug of syslog try to ping ppp client: slavon ~ # ping 192.168.0.2 PING 192.168.0.2 (192.168.0.2) 56(84) bytes of data. ping: sendmsg: No buffer space available ping: sendmsg: No buffer space available ping: sendmsg: No buffer space available ping: sendmsg: No buffer space available ping: sendmsg: No buffer space available ping: sendmsg: No buffer space available ping: sendmsg: No buffer space available ping: sendmsg: No buffer space available ping: sendmsg: No buffer space available ping: sendmsg: No buffer space available ping: sendmsg: No buffer space available ping: sendmsg: No buffer space available ping: sendmsg: No buffer space available #ifconfig ppp0 ppp0 Link encap:Point-to-Point Protocol inet addr:192.168.0.1 P-t-P:192.168.0.2 Mask:255.255.255.255 UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1396 Metric:1 RX packets:1648 errors:0 dropped:0 overruns:0 frame:0 TX packets:2407 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:3 RX bytes:75765 (73.9 Kb) TX bytes:3242633 (3.0 Mb) #iptables-save *nat :PREROUTING ACCEPT [6826:411515] :POSTROUTING ACCEPT [11117:531811] :OUTPUT ACCEPT [11118:531895] -A POSTROUTING -s 192.168.0.0/255.255.255.0 -j SNAT --to-source XXX.XXX.XXX.XXX COMMIT slavon ~ # pppd --version pppd version 2.4.3 slavon ~ # pptpd --version pptpd v1.3.1 I try it in 3 Computers 2 - x86 and 1 - x86-64 - All have this trouble its gentoo patches for kernel * Applying 1000_linux-2.6.16.1.patch (-p0+) * Applying 1001_linux-2.6.16.2.patch (-p0+) * Applying 1002_linux-2.6.16.3.patch (-p0+) * Applying 1003_linux-2.6.16.4.patch (-p0+) * Applying 1700_sparc-obp64-naming.patch (-p0+) * Applying 1705_sparc-U1-hme-lockup.patch (-p0+) * Applying 1710_sparc-sparsemem.patch (-p0+) * Applying 1715_ppc32-crosscc-ssp.patch (-p0+) * Applying 2400_sky2-1.1.patch (-p0+) * Applying 2405_skge-1.5.patch (-p0+) * Applying 2600_input-mod-devicetable.patch (-p0+) * Applying 4000_deprecate-sk98lin.patch (-p0+) * Applying 4100_libata-enable-atapi.patch (-p0+) * Applying 4105_dm-bbr.patch (-p0+) * Applying 4110_promise-pdc2037x.patch (-p0+) * Applying 4200_fbsplash-0.9.2-r5.patch (-p0+) * Applying 4205_vesafb-tng-1.0-rc1-r3.patch (-p0+) * Applying 4300_squashfs-3.0.patch (-p0+) * Applying 4400_speakup-20060103.patch (-p0+) * Applying 4401_speakup-highlight-tracking-amd64.patch (-p0+) * Applying 4405_alpha-sysctl-uac.patch (-p0+) !! In 2.6.15 kernel ALL WORK. ==================== I try Download ftp://ftp.altlinux.ru/pub/distributions/ALTLinux/ISO/Master-2.4-install-cd.iso after 5-10 MB network not responsible. Now try KERNEL 2.6.16-r3 -- bug is still here =) lsmod: Module Size Used by sha1 1992 - arc4 1480 - ppp_mppe 6092 - ppp_async 8968 - ppp_generic 25756 - slhc 5832 - crc_ccitt 1576 - ipt_DSCP 1704 - iptable_mangle 1896 - iptable_nat 6220 - ip_nat 14380 - ip_conntrack 42712 - nfnetlink 4736 - xt_tcpudp 2888 - iptable_filter 1960 - ip_tables 10228 - x_tables 9836 - usbhid 36448 - hw_random 4416 - ata_piix 7980 - libata 55188 - scsi_mod 84488 - uhci_hcd 29656 - ehci_hcd 28464 - usbcore 114820 -
Is this reproducible on the latest development kernel (2.6.17-rc2)? If so, do you have time to do a bisection? This will involve testing approximately 12 kernels between 2.6.15 and 2.6.16 (the last 5 or so will be very quick compiles). It will locate the *exact* patch which introduced the bug. http://www.reactivated.net/weblog/archives/2006/01/using-git-bisect-to-find-buggy-kernel-patches/
I try native 2.6.16.9 - bug still here. If u give me full patch in 1 file to 2.6.16.9 -> 2.6.17-rc2 - i test it. I can't test all patches 2.6.15 > 2.6.16 - its very many =)
(In reply to comment #4) > If u give me full patch in 1 file to 2.6.16.9 -> 2.6.17-rc2 - i test it. To go from *vanilla* 2.6.16.9 to 2.6.17-rc2 you can do: cd /usr/src/linux-2.6.16.9 bzcat patch-2.6.16.9.bz2 | patch -p1 -R bzcat patch-2.6.17-rc2.bz2 | patch -p1 The files you need are at: http://www.kernel.org/pub/linux/kernel/v2.6/patch-2.6.16.9.bz2 http://www.kernel.org/pub/linux/kernel/v2.6/testing/patch-2.6.17-rc2.bz2 > I can't test all patches 2.6.15 > 2.6.16 - its very many =) That's not what I asked - the "git bisect" method does what is called a binary search. So even though there were approximately 5000 patches inbetween 2.6.15 and 2.6.15, you'll only have to test about 13 kernels to find the offending patch. Still, if that is too much work, its understandable, but I'm not sure if this bug will go anywhere fast otherwise. Remember to test 2.6.17-rc2 and report back here before starting the bisection, if you choose to do that.
will track upstream bug
patch available upstream
.
Fixed in gentoo-sources-2.6.16-r10
