Description of problem: There appears the following on every DSL shutdown. Jun 10 19:04:33 server pppoe[1378]: read (syncReadFromPPP): Session 7622: Input/ output error Jun 10 19:04:33 server pppoe[1378]: Sent PADT Jun 10 19:04:33 server kernel: Badness in local_bh_enable at kernel/softirq.c:13 6 Jun 10 19:04:33 server kernel: Call Trace: Jun 10 19:04:33 server kernel: [<0211afe4>] local_bh_enable+0x39/0x56 Jun 10 19:04:33 server kernel: [<22b248fd>] ppp_sync_push+0xaf/0x115 [ppp_synct ty] Jun 10 19:04:33 server kernel: [<22b2443f>] ppp_sync_wakeup+0x1c/0x34 [ppp_sync tty] Jun 10 19:04:33 server kernel: [<021bb50c>] do_tty_hangup+0xeb/0x2d2 Jun 10 19:04:33 server kernel: [<021bc41e>] release_dev+0x1bc/0x4c0 Jun 10 19:04:33 server kernel: [<02134fd4>] zap_pmd_range+0x42/0x50 Jun 10 19:04:33 server kernel: [<0213501e>] unmap_page_range+0x3c/0x57 Jun 10 19:04:33 server kernel: [<021350fd>] unmap_vmas+0xc4/0x176 Jun 10 19:04:33 server kernel: [<021bc9fc>] tty_release+0x14/0x2d Jun 10 19:04:33 server adsl-stop: Killing adsl-connect Jun 10 19:04:33 server kernel: [<02141fa7>] __fput+0x3f/0xd1 Jun 10 19:04:33 server kernel: [<02140d99>] filp_close+0x59/0x5f Jun 10 19:04:33 server network: Shutting down interface ppp0: succeeded Jun 10 19:04:33 server kernel: [<021193ef>] put_files_struct+0x56/0xa9 Jun 10 19:04:33 server kernel: [<02119de0>] do_exit+0x1b3/0x2e3 Jun 10 19:04:33 server kernel: [<02119faa>] sys_exit_group+0x0/0xd Version-Release number of selected component (if applicable): kernel-2.6.5-1.358 How reproducible: Always. Steps to Reproduce: 1. Start DSL. 2. Stop DSL. 3. Watch logs. Actual results: There appears Badness. Expected results: No badness in log. Additional info:
I have the same problem using DSL with a clean FC2 install on two different machines (Microtel SYSMAR561 and Emachines 333cs). It persists with the 427 and 435 kernels.
I get the same thing - but to screen (on shutdown). Which log are you looking at?
When the xDSL network connection goes down automatically on shutdown, it only shows on the screen. However, if you bring the network connection down manually with Network Device Control without shutting down, the errors are written to /var/log/messages.
I found it in /var/log/messages after shutdown not manually bringing down. Here is corresponding line from syslog.conf: *.info;mail.none;authpriv.none;cron.none /var/log/messages
Strange. I have the same line in syslog.conf, but my errors are never written to /var/log/messages unless I bring down the connection manually.
Using kernel-2.6.7-1.494.2.2, the error appears to be gone. (BTW, I commented in bug #126190 that it appeared to be a dupe of this bug but there was no response.)