Bugzilla – Bug 6616
Complain about platform/sun4u/dscp:default in 2009.0906 sparc AI install in sun4v
Last modified: 2009-08-05 10:17:32 UTC
You need to log in before you can comment on or make changes to this bug.
This is found during testing sparc AI in sun4v with osol-0906-107-ai-sparc.iso. This is on line1-t2000 which is a sun4v system: Rebooting with command: boot net:dhcp Boot device: /pci@780/pci@0/pci@1/network@0:dhcp File and args: 1000 Mbps full duplex Link up 1000 Mbps full duplex Link up <time unavailable> wanboot info: WAN boot messages->console <time unavailable> wanboot info: Starting DHCP configuration <time unavailable> wanboot info: DHCP configuration succeeded <time unavailable> wanboot progress: wanbootfs: Read 366 of 366 kB (100%) <time unavailable> wanboot info: wanbootfs: Download complete Thu Feb 12 23:49:29 wanboot progress: miniroot: Read 170243 of 170243 kB (100%) Thu Feb 12 23:49:29 wanboot info: miniroot: Download complete SunOS Release 5.11 Version snv_106 64-bit Copyright 1983-2008 Sun Microsystems, Inc. All rights reserved. Use is subject to license terms. Hostname: opensolaris Remounting root read/write Probing for device nodes ... Preparing automated install image for use Downloading solaris.zlib archive --15:41:59-- http://10.6.68.21:5555/export/home/images/osol-0906-107-ai-sparc/solaris.zlib => `/tmp/solaris.zlib' Connecting to 10.6.68.21:5555... connected. HTTP request sent, awaiting response... 200 OK Length: 324,127,744 (309M) [text/plain] 11% [===> ] 37,891,760 22.16M/s 15:42:01 (22.13 MB/s) - Read error at byte 37891760/324127744 (Connection reset by peer). Retrying. --15:42:02-- http://10.6.68.21:5555/export/home/images/osol-0906-107-ai-sparc/solaris.zlib (try: 2) => `/tmp/solaris.zlib' Connecting to 10.6.68.21:5555... connected. HTTP request sent, awaiting response... 206 Partial Content Length: 324,127,744 (309M), 286,235,984 (273M) remaining [text/plain] 48% [++++=============> ] 157,724,336 38.23M/s 15:42:05 (38.17 MB/s) - Read error at byte 157724336/324127744 (Connection reset by peer). Retrying. --15:42:07-- http://10.6.68.21:5555/export/home/images/osol-0906-107-ai-sparc/solaris.zlib (try: 3) => `/tmp/solaris.zlib' Connecting to 10.6.68.21:5555... connected. HTTP request sent, awaiting response... 206 Partial Content Length: 324,127,744 (309M), 166,403,408 (159M) remaining [text/plain] 70% [++++++++++++++++++=======> ] 229,797,616 8.37M/s ETA 00:25 15:42:22 (4.75 MB/s) - Read error at byte 229797616/324127744 (Connection reset by peer). Retrying. --15:42:25-- http://10.6.68.21:5555/export/home/images/osol-0906-107-ai-sparc/solaris.zlib (try: 4) => `/tmp/solaris.zlib' Connecting to 10.6.68.21:5555... connected. HTTP request sent, awaiting response... 206 Partial Content Length: 324,127,744 (309M), 94,330,128 (90M) remaining [text/plain] 100%[++++++++++++++++++++++++++==========>] 324,127,744 21.46M/s ETA 00:00 15:42:40 (6.01 MB/s) - `/tmp/solaris.zlib' saved [324127744/324127744] Downloading solarismisc.zlib archive --15:42:40-- http://10.6.68.21:5555/export/home/images/osol-0906-107-ai-sparc/solarismisc.zlib => `/tmp/solarismisc.zlib' Connecting to 10.6.68.21:5555... connected. HTTP request sent, awaiting response... 200 OK Length: 913,920 (892K) [text/plain] 100%[====================================>] 913,920 --.--K/s 15:42:40 (37.05 MB/s) - `/tmp/solarismisc.zlib' saved [913920/913920] --15:42:40-- http://10.6.68.21:5555/export/home/images/osol-0906-107-ai-sparc/install.conf => `/tmp/install.conf' Connecting to 10.6.68.21:5555... connected. HTTP request sent, awaiting response... 200 OK Length: 39 [text/plain] 100%[====================================>] 39 --.--K/s 15:42:40 (893.03 KB/s) - `/tmp/install.conf' saved [39/39] Done mounting automated install image Configuring devices. Reading ZFS config: done. Feb 12 15:43:01 svc.startd[7]: svc:/platform/sun4u/dscp:default: Method "/lib/svc/method/svc-dscp start" failed with exit status 96. Feb 12 15:43:01 svc.startd[7]: platform/sun4u/dscp:default misconfigured: transitioned to maintenance (see 'svcs -xv' for details) opensolaris console login: root password: root@opensolaris:/tmp# more /var/svc/log/platform-sun4u-dscp:default.log [ Feb 12 15:43:01 Executing start method ("/lib/svc/method/svc-dscp start"). ] [ Feb 12 15:43:01 Method "start" exited with status 96. ] root@opensolaris:/tmp#
Assigning to Sundar for additional triage within AI team.
Recieved this installation failure message for Chicago sun4u Feb 19 16:07:43 svc.startd[7]: svc:/platform/sun4u/dscp:default: Method "/lib/svc/method/svc-dscp start" failed with exit status 96. Feb 19 16:07:43 svc.startd[7]: platform/sun4u/dscp:default misconfigured: transitioned to maintenance (see 'svcs -xv' for details)
Actually, it's enabled by default on all sparcs. However, first thing what start script does is checking is it OPL or not. If not, it fails.
for 2009.06 I will be removing this SUNWdscp from AI image.
If we want to remove SUNWdcsp, we would also have to remove SUNWfmd which depends on that one and which (if present in AI DC manifest) would pull SUNWdscp into AI image: # pkg uninstall -nv SUNWdscp Creating Plan - Before evaluation: UNEVALUATED: -pkg:/SUNWdscp@0.5.11,5.11-0.111:20090418T184415Z Creating Plan /pkg: Cannot remove 'pkg:/SUNWdscp@0.5.11,5.11-0.111:20090418T184415Z' due to the following packages that depend on it: pkg:/SUNWfmd@0.5.11,5.11-0.111:20090418T184942Z # pkg info SUNWfmd Name: SUNWfmd Summary: Fault Management Daemon and Utilities Category: System/Administration and Configuration State: Installed Publisher: opensolaris.org Version: 0.5.11 Build Release: 5.11 Branch: 0.111 Packaging Date: Sat Apr 18 18:49:42 2009 Size: 8.69 MB FMRI: pkg:/SUNWfmd@0.5.11,5.11-0.111:20090418T184942Z
Why is this service enabled by default for all sparcs? In NV at least, it looks to be enabled only by the platform_SUNW,SPARC-Enterprise.xml profile. I would think the same should hold true for OpenSolaris.
It looks to be an issue in our image build process. We decide what the /var/svc/profile/platform.xml points to in the DC's mkrepo script. That script appears to use the arch of the running build system to pick what platform.xml points to: # 3. Profile application. We must create the platform profile upon # first boot, as we may be a diskless client of a platform or # architecture distinct from our NFS server. # svccfg_apply ${ROOTDIR}/var/svc/profile/generic_limited_net.xml if [ ! -f ${ROOTDIR}/var/svc/profile/platform.xml ]; then this_karch=`uname -m` this_plat=`uname -i` if [ -f ${ROOTDIR}/var/svc/profile/platform_$this_plat.xml ]; then platform_profile=platform_$this_plat.xml elif [ -f ${ROOTDIR}/var/svc/profile/platform_$this_karch.xml ]; then platform_profile=platform_$this_karch.xml else platform_profile=platform_none.xml fi (cd ${ROOTDIR}/var/svc/profile; ln -s $platform_profile platform.xml) fi So this is probably why this bug, along with bug 7683 and bug 6578 exist.
*** Bug 6578 has been marked as a duplicate of this bug. ***
*** Bug 7683 has been marked as a duplicate of this bug. ***
After further digging around, the root cause appears to be that the correct platform profile is not being applied and hence these errors occur. The better approach would be to move manifest import functionality from the image construction phase to the first boot phase. Such a change is significant and needs to be better understood. Comments indicate that these failure do not prevent installing the system. Since we are closing the gate for 2009.06 any day now, these failures will be release noted. I am removing this from the blocker list.
Bug 5209 is the bug for "improving/removing" the mkrepo script. The solution for fixing that bug should address take into account of the problem reported here.
This is true not only for dscp but also for the sckmd service. The following are console messages from AI install of OpenSolaris 2009.06 image on a T5120 machine: Done mounting automated install image Configuring devices. Reading ZFS config: done. Jun 18 07:00:51 svc.startd[7]: svc:/platform/sun4u/sckmd:default: Method "/lib/svc/method/svc-sckmd" failed with exit status 98. Jun 18 07:00:51 svc.startd[7]: svc:/platform/sun4u/sckmd:default: Method "/lib/svc/method/svc-sckmd" failed with exit status 98. Jun 18 07:00:52 svc.startd[7]: svc:/platform/sun4u/sckmd:default: Method "/lib/svc/method/svc-sckmd" failed with exit status 98. Jun 18 07:00:52 svc.startd[7]: platform/sun4u/sckmd:default failed: transitioned to maintenance (see 'svcs -xv' for details) Jun 18 07:00:52 svc.startd[7]: svc:/platform/sun4u/dscp:default: Method "/lib/svc/method/svc-dscp start" failed with exit status 96. Jun 18 07:00:52 svc.startd[7]: platform/sun4u/dscp:default misconfigured: transitioned to maintenance (see 'svcs -xv' for details) Service discovery phase initiated Service name to look up: sparc_0906_installsvc
This bug needs to be looked in context of building images by the distro-constructor.
When the work for 5209 is completed, this problem will be taken care of. *** This bug has been marked as a duplicate of bug 5209 ***