Bug 378641
Summary: | SIGABRT when trying to get an IP address in loader | ||||||||
---|---|---|---|---|---|---|---|---|---|
Product: | [Fedora] Fedora | Reporter: | Jesse Keating <jkeating> | ||||||
Component: | anaconda | Assignee: | David Cantrell <dcantrell> | ||||||
Status: | CLOSED RAWHIDE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||||
Severity: | low | Docs Contact: | |||||||
Priority: | low | ||||||||
Version: | rawhide | CC: | bpeck, chrisw, dcantrell, jlaska, lkundrak, michal, poelstra, redwolfe, wwoods | ||||||
Target Milestone: | --- | Keywords: | Reopened | ||||||
Target Release: | --- | ||||||||
Hardware: | x86_64 | ||||||||
OS: | Linux | ||||||||
Whiteboard: | |||||||||
Fixed In Version: | dhcp-4.0.0-6.fc9 | Doc Type: | Bug Fix | ||||||
Doc Text: | Story Points: | --- | |||||||
Clone Of: | Environment: | ||||||||
Last Closed: | 2008-01-28 22:08:14 UTC | Type: | --- | ||||||
Regression: | --- | Mount Type: | --- | ||||||
Documentation: | --- | CRM: | |||||||
Verified Versions: | Category: | --- | |||||||
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||
Cloudforms Team: | --- | Target Upstream Version: | |||||||
Embargoed: | |||||||||
Bug Depends On: | |||||||||
Bug Blocks: | 428703 | ||||||||
Attachments: |
|
Description
Jesse Keating
2007-11-12 19:54:57 UTC
I need the anaconda debuginfo package for the tree you booted in order to debug this. *** Bug 418181 has been marked as a duplicate of this bug. *** Fixed in libdhcp-1.99.4. The libnl API changed a bit. The anaconda-11.4.0.15-1 build is linked with libdhcp-1.99.4, so it should be working fine. I've tested here and all seems well. *** Bug 390611 has been marked as a duplicate of this bug. *** *** Bug 404691 has been marked as a duplicate of this bug. *** *** Bug 426624 has been marked as a duplicate of this bug. *** I'm re-opening this. It seems to still be happening on x86_64. I'm attaching a photo of the traceback (which is useful now!) Created attachment 290851 [details]
Photo of traceback
Confirmed that this only happens on x86_64. i386 gets past this (and tracebacks somewhere else). Still happening here w/ today's rawhide (Jan 9), with same back trace that jkeating reported: loader received SIGABRT! Backtrace: /sbin/loader(loaderSegvHandler+0x7e)[0x409ace] /lib64/libc.so.6[0x2aaaacd5d000] /lib64/libc.so.6(gsignal+0x35)[0x2aaaacd5cf95] /lib64/libc.so.6(abort+0x110)[0x2aaaacd5ea40] /lib64/libc.so.6[0x2aaaacda4fac] /lib64/libc.so.6(cfree+0x27)[0x2aaaacda4777] /lib64/libc.so.6[0x2aaaace012ee] /lib64/libc.so.6(tdestroy+0x38)[0x2aaaace013d8] /lib64/libdhcp-1.99.so.1(dhcpv4_lease_free+0x5b)[0x2aaaab91a5ab] /lib64/libdhcp-1.99.so.1(dhcpv4_control_free+0x1a)[0x2aaaab91c03a] /lib64/libdhcp-1.99.so.1(dhcp_nic_free+0x11)[0x2aaaab91db31] /lib64/libdhcp-1.99.so.1(pumpSetupInterface+0x66)[0x2aaaab925416] /sbin/loader(configureNetwork+0x11)[0x41ab31] /sbin/loader(readNetConfig+0x53f)[0x41cfaf] /sbin/loader(main+0xade)[0x40adce] /lib64/libc.so.6(__libc_start_main+0xf4)[0x2aaaacd492b4] /sbin/loader[0x4081f9] install exited abnormally [1/1] *** Bug 428227 has been marked as a duplicate of this bug. *** Fixed this in rawhide in libdhcp-1.99.6-1.fc9. Building new packages now. Still failing with rawhide-20080116 which has libdhcp-1.99.6-1.fc9.src.rpm. Logs follow for i386 now. loader received SIGSEGV! Backtrace: /sbin/loader(loaderSegvHandler+0xa0)[0x8050060] [0x12f420] /lib/libc.so.6(fclose+0x1d)[0x54fbbd] /lib/libdhcp4client-4.0.so.0(end_iface_scan+0x25)[0x3b4975] /lib/libdhcp4client-4.0.so.0(discover_interfaces+0x2b7)[0x3b5e47] /lib/libdhcp4client-4.0.so.0(dhcpv4_client+0x111f)[0x3a57df] /lib/libdhcp-1.99.so.1(do_dhcpv4+0x3c)[0x32dbbc] /lib/libdhcp-1.99.so.1[0x32f975] /lib/libdhcp-1.99.so.1(dhcp_nic+0x3a)[0x32feea] /lib/libdhcp-1.99.so.1(pumpDhcpClassRun+0x127)[0x337ca7] /sbin/loader(doDhcp+0x12a)[0x8062c1a] /sbin/loader(configureTCPIP+0x9b8)[0x80653d8] /sbin/loader(readNetConfig+0x3bd)[0x8065aed] /sbin/loader(main+0xd9a)[0x805175a] /lib/libc.so.6(__libc_start_main+0xe0)[0x5094a0] /sbin/loader[0x804e121] install exited abnormally [1/1] sending termination signals...done sending kill signals...done disabling swap... unmounting filesystems... /proc done /dev/pts done /sys done you may safely reboot your system loader received SIGABRT! Backtrace: /sbin/loader(loaderSegvHandler+0xa0)[0x8050060] [0x12f420] /lib/libc.so.6(gsignal+0x46)[0x51da66] /lib/libc.so.6(abort+0x101)[0x51f421] /lib/libc.so.6[0x5658a5] /lib/libc.so.6(cfree+0x35)[0x5650b5] /lib/libdhcp4client-4.0.so.0(dfree+0x26)[0x3e5966] /lib/libdhcp4client-4.0.so.0(script_init+0x46)[0x3a2126] /lib/libdhcp4client-4.0.so.0(bind_lease+0x5a)[0x3a37da] /lib/libdhcp4client-4.0.so.0(state_selecting+0xca)[0x3a400a] /lib/libdhcp4client-4.0.so.0(process_outstanding_timeouts+0xf8)[0x3b6a98] /lib/libdhcp4client-4.0.so.0(dhcpv4_client+0x1777)[0x3a5e37] /lib/libdhcp-1.99.so.1(do_dhcpv4+0x3c)[0x32dbbc] /lib/libdhcp-1.99.so.1[0x32f975] /lib/libdhcp-1.99.so.1(dhcp_nic+0x3a)[0x32feea] /lib/libdhcp-1.99.so.1(pumpDhcpClassRun+0x127)[0x337ca7] /sbin/loader(doDhcp+0x12a)[0x8062c1a] /sbin/loader(configureTCPIP+0x9b8)[0x80653d8] /sbin/loader(readNetConfig+0x3bd)[0x8065aed] /sbin/loader(main+0xd9a)[0x805175a] /lib/libc.so.6(__libc_start_main+0xe0)[0x5094a0] /sbin/loader[0x804e121] install exited abnormally [1/1] sending termination signals...done sending kill signals...done disabling swap... unmounting filesystems... /proc done /dev/pts done /sys done you may safely reboot your system According to bpeck, tested against rawhide-20080122-Default and still broken How's bpeck triggering the bug? especially on i386? rawhide-20080122 works fine for me. I'm doing an nfs install with dhcp. "ip=dhcp ipv6=auto" as kernel boot args? or requesting dhcp interactively in loader? Worksforme with 2008-01-22 images. With everything available over NFS and network interfaces configured via dhcp from _within_ anaconda I can use those images at least to reach a shell prompt while booting "rescue". Created attachment 292703 [details]
tail of install logs showing SIGABRT
parameters passeed via pxe are console and ks.
kickstart file has network --device eth0 --bootproto dhcp
One thing to note is I can see anaconda pulling the kickstart file from the
server via http. Its after it pulls the kickstart file that this blows up. So
its definitely kickstart related.
I have had some other systems proceed past this point but they blow up at
different points. I've filed BZ's on them as well. The only way I can get a
successful install is to disable selinux or not specify it in the kickstart
That trace runs through getFileFromUrl, so (in theory) you're done with DHCP and you've got an IP at this point. So, yes, it's a SIGABRT in loader, but this is a different bug. This should be fixed with dhcp-4.0.0-6.fc9, which is tagged for the beta. |