-unstable

-unstable

[Ian Pratt]

The work over the weekend to cleanup xenbus seems to have exposed a
latent bug in the block device code, the net result being that the
current unstable tree struggles to start guests with block devices on
several of our systems.
If you're doing tests please use changeset 7267:e69413dca684 (Fix
locking in /dev/xen/evtchn) which is now tagged 'semi-stable' in the
repo.

We expect this issue to be fixed in a few hours.

Ian

Re: -unstable

[Chris Bainbridge]

On 10/10/05, Ian Pratt <m+Ian.Pratt@cl.cam.ac.uk> wrote:
>
> If you're doing tests please use changeset 7267:e69413dca684 (Fix
> locking in /dev/xen/evtchn) which is now tagged 'semi-stable' in the
> repo.

With 7267 I'm often seeing these errors starting xend:

[2005-10-10 13:31:45 xend] INFO (SrvDaemon:301) Xend Daemon started
[2005-10-10 13:31:45 xend] DEBUG (XendDomainInfo:174)
XendDomainInfo.recreate({'vcpu_to_cpu': [0, 1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1], 'cpu_time': 31060300300L, 'ssidref': 0,
'shutdown_reason': 0, 'dying': 0, 'dom': 0, 'running': 1, 'maxmem_kb':
-4, 'paused': 0, 'crashed': 0, 'vcpus': 2, 'shutdown': 0, 'mem_kb':
93184, 'cpumap': [-1, -1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], 'blocked': 0})
[2005-10-10 13:31:45 xend] INFO (XendDomainInfo:193) Recreating domain
0, UUID 504408b2-5c64-449e-87bc-aabb3d2d333c.
[2005-10-10 13:31:45 xend] ERROR (XendDomain:183) Failed to recreate
information for domain 0.  Doing nothing except crossing my fingers.
Traceback (most recent call last):
  File "/usr/lib/python2.4/site-packages/xen/xend/XendDomain.py", line
176, in refresh
    dominfo = XendDomainInfo.recreate(doms[d])
  File "/usr/lib/python2.4/site-packages/xen/xend/XendDomainInfo.py",
line 211, in recreate
    vm.initStoreConnection()
  File "/usr/lib/python2.4/site-packages/xen/xend/XendDomainInfo.py",
line 1379, in initStoreConnection
    self.dompath)
  File "/usr/lib/python2.4/site-packages/xen/xend/xenstore/xsutil.py",
line 20, in IntroduceDomain
    return xshandle().introduce_domain(domid, page, port, path)
RuntimeError: (22, 'Invalid argument')
[2005-10-10 13:31:45 xend] ERROR (SrvDaemon:311) Exception starting xend (0)
Traceback (most recent call last):
  File "/usr/lib/python2.4/site-packages/xen/xend/server/SrvDaemon.py",
line 304, in run
    servers = SrvServer.create()
  File "/usr/lib/python2.4/site-packages/xen/xend/server/SrvServer.py",
line 99, in create
    root.putChild('xend', SrvRoot())
  File "/usr/lib/python2.4/site-packages/xen/xend/server/SrvRoot.py",
line 40, in __init__
    self.get(name)
  File "/usr/lib/python2.4/site-packages/xen/web/SrvDir.py", line 82, in get
    val = val.getobj()
  File "/usr/lib/python2.4/site-packages/xen/web/SrvDir.py", line 52, in getobj
    self.obj = klassobj()
  File "/usr/lib/python2.4/site-packages/xen/xend/server/SrvDomainDir.py",
line 38, in __init__
    self.xd = XendDomain.instance()
  File "/usr/lib/python2.4/site-packages/xen/xend/XendDomain.py", line
586, in instance
    inst = XendDomain()
  File "/usr/lib/python2.4/site-packages/xen/xend/XendDomain.py", line
66, in __init__
    self.dom0_setup()
  File "/usr/lib/python2.4/site-packages/xen/xend/XendDomain.py", line
131, in dom0_setup
    dom0 = self.domains[PRIV_DOMAIN]
KeyError: 0

Re: -unstable

[Charles Duffy]

Ian Pratt wrote:
> If you're doing tests please use changeset 7267:e69413dca684 (Fix
> locking in /dev/xen/evtchn) which is now tagged 'semi-stable' in the
> repo.

I've just done a pull, and see a "semistable" tag on 2327:9afec5bc14ae, 
but none on 7267:e69413dca684.

Indeed, my full set of tags is as follows:

coruscant:~/xen-unstable.hg # hg tags
tip               7284:1ac39c7a043541cfa94655f0e9ab98d4503c29a2
ia64-stable       5606:6a65fe0f84c8339b5b89362d0ec34d8abab752b0
RELEASE-2.0.6     5080:2a5814ad2e5634a5fa291b703a152e7fc0b4faf0
RELEASE-2.0.5     4082:1a522944f76540ea9d73fcc1b0d13d0f670183f0
RELEASE-2.0.4     3636:fb875591fd72e15c31879c0e9034d99b80225595
RELEASE-2.0.3     3416:6e1bbc13911751efa0b1c018425c1b085820fa02
RELEASE-2.0.2     3307:dc2f08429f17e6614fd2f1ab88cc09ca0a850f32
RELEASE-2.0.1     3019:475a162b66e2c19b1e9468b234a4ba705334905e
RELEASE-2.0.0     2905:42882b3e0dda89f3a8ec00da568f86e9b3c230f1
semistable        2327:9afec5bc14aeb197ef37ea54a57eacd427463fc3
split-1.1         858:3d330e41f41ce1bc118c02346e18949ad5d67f6b
latest-semistable 858:3d330e41f41ce1bc118c02346e18949ad5d67f6b
split-1.0         779:30c521db4c71960b0cf1d9c9e1b658e77b535a3e
latest-stable     779:30c521db4c71960b0cf1d9c9e1b658e77b535a3e
beta1             454:487b2ee37d1cecb5f3e7a546b05ad097a0226f2f

Re: -unstable

[Keir Fraser]

On 10 Oct 2005, at 11:38, Ian Pratt wrote:

> The work over the weekend to cleanup xenbus seems to have exposed a
> latent bug in the block device code, the net result being that the
> current unstable tree struggles to start guests with block devices on
> several of our systems.
> If you're doing tests please use changeset 7267:e69413dca684 (Fix
> locking in /dev/xen/evtchn) which is now tagged 'semi-stable' in the
> repo.
>
> We expect this issue to be fixed in a few hours.

Not a latent bug after all, but it should be fixed now. :-)

  -- Keir