On 2018-01-14 02:15:41, Iustin Pop wrote:
> Hi all,
>
> I did a quick look at the GitHub issue tracker today, and it seems that
> the migrator did not close the migrated issues that were closed.
>
> More specifically, a search of:
>
> is:issue is:open label:St
On 2018-02-04 14:18:51, Ganeti Development List wrote:
> As an FYI all, I have closed all Github issues that had the label
> Status:Fixed.
Awesome, thanks Morg!
iustin
Hi all,
I did a quick look at the GitHub issue tracker today, and it seems that
the migrator did not close the migrated issues that were closed.
More specifically, a search of:
is:issue is:open label:Status:Fixed
Shows 98 such open issues. If I'm not mistaken, "Status:Fixed" is a
Hi Apollon,
I just realised that all the fixes from the Debian packaging are on
master. This means two thing for the stable-2.16 branch:
- building in a recent environment is tricky
- the travis-ci builds for it are still failing
Any reason why not to update stable-2.16 to be fully buildable?
On 2019-02-12 17:02:35, Apollon Oikonomopoulos wrote:
> Hi everyone,
One specific question below:
> - git push --force origin v2.16.0:master and continue development from
>there. I don't expect the force-push to be an issue, and it's much
>cleaner than a `git merge -s ours' or messing
And now correcting the -devel mailing list address.
On 2019-02-12 18:26:04, Iustin Pop wrote:
> On 2019-02-12 17:02:35, Apollon Oikonomopoulos wrote:
> > Hi everyone,
>
> Hi Apollon,
>
> > With 2.16.0 released last fall, I think it's time to discuss how we can
> >
On 2019-02-13 00:56:49, Apollon Oikonomopoulos wrote:
> On 18:38 Tue 12 Feb , Iustin Pop wrote:
> > On 2019-02-12 17:02:35, Apollon Oikonomopoulos wrote:
> > > Hi everyone,
> >
> > One specific question below:
> >
> > > - git push --force origin
On 2019-02-12 17:02:35, Apollon Oikonomopoulos wrote:
> Immediate steps
> ---
>
> In light of the above, I propose the following:
>
> - Dismiss all unreleased work and archive the current master and
>stable-2.17 branches under attic/ (I've already pushed a ref for
>
Hi all,
I made the mistake of attempting to build ganeti 3.0 beta on a sid box,
with GHC 8.8, which causes no end of fun…
To solve the fun and make this actually built, would be good to settle
which versions of GHC we support (if we didn't already and I missed the
email or simply forgot).
IMO,
/1492
Ouch, I missed those.
I have Ganeti now working with GHC 8.8, but cleaning up the patches
depends on what the minimum version will be.
Thanks, I'll reply on those bugs.
iustin
> On Tue, Jul 7, 2020 at 11:20 PM Iustin Pop wrote:
>
> > Hi all,
> >
> > I made the m
uld keep 8.0 compatibility.
It's doable, but not pleasant :)
I'll reply on bug too.
iustin
> Am Mi., 8. Juli 2020 um 10:19 Uhr schrieb Iustin Pop :
> >
> > On 2020-07-08 08:22:22, Rudolph Bott wrote:
> > > Hi Iustin,
> > >
> > > I can not add anythi
Just FYI, as I enabled one of my own repositories for the security scan
beta, I realised I can enable ganeti too, so I went ahead and did that
for the entire org.
It's s closed beta, so we're in the queue only, not yet enabled, we'll
see how it goes.
iustin
--
You received this message because
On 2020-12-01 18:30:45, Apollon Oikonomopoulos wrote:
> Hi Sascha,
>
> Apologies for not responding earlier, end-November is always terribly
> busy.
Same here. Especially as Debian freeze approaches…
> Sascha Lucas writes:
>
> > Hi,
> >
> > after some weeks of silence I like to ask if someone
On 2020-12-30 11:50:10, Apollon Oikonomopoulos wrote:
> Hi all,
>
> Now that we've officially released 3.0, it's time to look a bit at
> stable release management. What I'd like to discuss in particular is the
> possibility of switching from a forward-porting workflow for stable
> updates to a
Basically: https://lwn.net/Articles/845480/. Examples/tutorial in
https://www.python.org/dev/peps/pep-0636/.
Don't know in which Python version will land, but it looks to improve
code safety quite a bit (in some cases). Then it needs to land in Debian
Stable, etc. etc. :)
iustin
--
You
On Thu, Aug 28, 2008 at 01:53:53PM +0200, Michael Hanselmann wrote:
HVM options were missing.
---
scripts/gnt-backup | 43 ++-
1 files changed, 42 insertions(+), 1 deletions(-)
diff --git a/scripts/gnt-backup b/scripts/gnt-backup
index
This patch adds a new '-s' parameter to ‘gnt-instance info’ that makes
it return only 'static' information. This is much faster, especially for
drbd instances.
---
lib/cmdlib.py | 26 +-
lib/opcodes.py|2 +-
man/gnt-instance.sgml | 12 +++-
On Thu, Aug 28, 2008 at 03:18:07PM +0200, Michael Hanselmann wrote:
On Thu, Aug 28, 2008 at 1:59 PM, Iustin Pop [EMAIL PROTECTED] wrote:
On Thu, Aug 28, 2008 at 01:53:52PM +0200, Michael Hanselmann wrote:
diff --git a/scripts/gnt-instance b/scripts/gnt-instance
index c03dc83..dbac540
On Thu, Aug 28, 2008 at 03:41:47PM +0200, Michael Hanselmann wrote:
HVM options were missing.
---
scripts/gnt-backup | 39 ++-
1 files changed, 38 insertions(+), 1 deletions(-)
LGTM.
This just adds the newly added field to LUQueryInstances into the RAPI
instance queries.
---
lib/rapi/resources.py | 56 ++--
1 files changed, 35 insertions(+), 21 deletions(-)
diff --git a/lib/rapi/resources.py b/lib/rapi/resources.py
index
On Thu, Aug 28, 2008 at 05:18:29PM +0200, Michael Hanselmann wrote:
2008/8/28 Michael Hanselmann [EMAIL PROTECTED]:
Reported by Iustin.
Interdiff:
LGTM, thanks!
iustin
On Fri, Aug 29, 2008 at 11:08:10AM +0100, Guido Trotter wrote:
On Fri, Aug 29, 2008 at 11:06 AM, Iustin Pop [EMAIL PROTECTED] wrote:
+prev_log_serial, timeou))
s/timeou/timeout/ :)
Argh... done.
iustin
On Fri, Aug 29, 2008 at 06:52:33AM -0700, Alexander Schreiber wrote:
Hello ganeti-devel@googlegroups.com,
I'd like you to do a code review on the patch below:
Merge r1295 from branches/ganeti/ganeti-1.2
Clarify cluster IP requirement.
Code location: .../ganeti/
LGTM, thanks.
On Mon, Sep 01, 2008 at 10:40:32AM +0200, Michael Hanselmann wrote:
2008/8/29 Iustin Pop [EMAIL PROTECTED]:
doc/examples/Makefile.am|2 +-
doc/examples/batcher-instances.json | 14 +
tools/Makefile.am |4 +-
tools/README.batcher
On Mon, Sep 01, 2008 at 10:52:54AM +0100, Guido Trotter wrote:
It was already allowed in gnt-instance modify, but ignored.
It will be used to force skipping parameter checks.
Reviewed-by: imsnah
git-svn-id: https://ganeti.googlecode.com/svn/branches/ganeti/[EMAIL
PROTECTED]
On Thu, Sep 04, 2008 at 03:36:52PM +0100, Guido Trotter wrote:
It turns out this fields was already mandatory. If it hadn't beed valid,
in fact, a value of None would have been passed to _GetWantedNodes which
would have thrown an exception.
---
lib/cmdlib.py |2 +-
1 files changed, 1
On Thu, Sep 04, 2008 at 03:36:55PM +0100, Guido Trotter wrote:
All the users of _GetWantedNodes have been converted to be concurrent
LUs, and thus cannot call this function with an empty list of nodes
anymore. This patch makes this restriction a part of the function
itself. This prevents
On Fri, Sep 05, 2008 at 11:18:55AM +0100, Guido Trotter wrote:
On Fri, Sep 5, 2008 at 11:15 AM, Michael Hanselmann [EMAIL PROTECTED] wrote:
SplitTime didn't round the same on different platforms. I've changed
it to more or less undo what the Python code does by dividing by
0.01.
On Mon, Sep 08, 2008 at 04:26:35PM +0100, Guido Trotter wrote:
Fixing the check in gnt-cluster, or gnt-cluster verify-disks is broken.
Since the version in 1.2 used to return a tuple we'll accept both.
---
LGTM.
On Mon, Sep 08, 2008 at 04:26:36PM +0100, Guido Trotter wrote:
The rpc library returns a list, not a tuple, so we'll accept both.
---
LGTM.
On Mon, Sep 08, 2008 at 03:51:26PM +0100, Guido Trotter wrote:
This function was incompatible with the new locking system, and its
usage has been removed from the code. For now LUs share code by calling
common module-private functions in cmdlib.py, in the future they will
use tasklets (when
On Mon, Sep 08, 2008 at 03:51:28PM +0100, Guido Trotter wrote:
This is the most complex parallelization so far. We have to lock one
instance (and its nodes) plus one more node if doing a remote replace,
or all nodes if doing a remote replace with iallocator.
---
lib/cmdlib.py | 54
On Tue, Sep 09, 2008 at 10:55:42AM +0100, Guido Trotter wrote:
This constant replaces what we used to write in recalculate_locks, and
represents the lock recalculation mode. It lives in constants.py because
it's used only in cmdlib, and thus doesn't deal with the locking library
by itself.
On Tue, Sep 09, 2008 at 10:55:43AM +0100, Guido Trotter wrote:
This will be used to lock the instance's nodes in addition to some more.
---
lib/cmdlib.py|6 +-
lib/constants.py |1 +
2 files changed, 6 insertions(+), 1 deletions(-)
LGTM.
On Tue, Sep 09, 2008 at 12:02:46PM +0100, Guido Trotter wrote:
Release a new Ganeti 1.2.6 candidate.
release candidate 1 includes:
- the NEWS file
- HVM backwards compatibility fixes
- disable LUGrowDisk in burnin, for diskless instances
- manpage fixes
- protocol version bump up
The _GetMasterInfo() function needs to export the master name too to be
useful in master safety checks. This patch makes it a public (no _)
function and adds a third element in the return tuple. Its callers are
modified too.
---
lib/backend.py | 17 -
1 files changed, 12
These small patches implement master startup checks via a (positive
negative) rule (except for one node clusters, where we always start).
Iustin Pop (3):
Change backend._GetMasterInfo to return more data
Export backend.GetMasterInfo over the rpc layer
Implement master startup safety check
On Tue, Sep 09, 2008 at 02:19:32PM +0200, Michael Hanselmann wrote:
On Tue, Sep 9, 2008 at 1:53 PM, Iustin Pop [EMAIL PROTECTED] wrote:
+def CheckAgreement():
[…]
+ results = rpc.call_master_info(node_list)
+ if not isinstance(results, dict):
+# this should not happen (unless
On Tue, Sep 09, 2008 at 02:31:16PM +0200, Michael Hanselmann wrote:
Otherwise, corruption could occur in some corner cases. E.g. when
LeaveNode is running in a child and is in the process of removing
queue files, the main process gets killed, started again and gets
a request to update the
On Tue, Sep 09, 2008 at 02:31:17PM +0200, Michael Hanselmann wrote:
This is a left-over from some experiments. It creates a temporary
file, creates a utils.FileLock instance for it and unlinks the file.
---
lib/utils.py | 16
1 files changed, 16 insertions(+), 0
On Tue, Sep 09, 2008 at 03:10:16PM +0100, Guido Trotter wrote:
Now that is_owned is public we don't need to play games at the end of an
LU. If we're still owning anything we just release it.
---
lib/mcpu.py |4 +---
1 files changed, 1 insertions(+), 3 deletions(-)
LGTM.
On Tue, Sep 09, 2008 at 03:57:35PM +0100, Guido Trotter wrote:
On Tue, Sep 9, 2008 at 3:49 PM, Iustin Pop [EMAIL PROTECTED] wrote:
On Tue, Sep 09, 2008 at 03:10:17PM +0100, Guido Trotter wrote:
With this patch LUs can declare locks to be added when they start and/or
removed after
On Tue, Sep 09, 2008 at 04:37:23PM +0100, Guido Trotter wrote:
On Tue, Sep 9, 2008 at 4:34 PM, Iustin Pop [EMAIL PROTECTED] wrote:
Well, how likely is an assignement to fail compared to a lock add to
fail? IMHO, assignements cannot fail in normal operations...
Highly unkilely, yes
The sys.exit(0) was not nice as you couldn't differentiate it from
other exists. We change this to a specially defined exception for this,
so that multi-opcode commands can handle this nicely.
---
lib/cli.py|7 +--
lib/errors.py |9 +
2 files changed, 14 insertions(+), 2
On Tue, Sep 09, 2008 at 05:53:46PM +0200, Michael Hanselmann wrote:
2008/9/9 Iustin Pop [EMAIL PROTECTED]:
The sys.exit(0) was not nice as you couldn't differentiate it from
other exists. We change this to a specially defined exception for this,
s/exists/exit codes/
so that multi
On Tue, Sep 09, 2008 at 06:40:08PM +0100, Guido Trotter wrote:
On Tue, Sep 9, 2008 at 4:42 PM, Iustin Pop [EMAIL PROTECTED] wrote:
s/we never/we today don't/.
We definitely don't ever want to, as add failing most probably denotes
a failed race condition with another job that already
On Tue, Sep 09, 2008 at 06:53:46PM +0100, Guido Trotter wrote:
I was asked to change this in such a way that if a = b failed due to
OOM I didn't forget to release the lock.
Hmm hmm... Michael, in what situation could we fail to assign a=b
but otherwise the program would continue to run?
This patch splits the GetInstanceInfo and GetInstanceList methods into
two parts, one locked one _Unlocked similar to the way nodes are
queried.
---
lib/config.py | 33 -
1 files changed, 24 insertions(+), 9 deletions(-)
diff --git a/lib/config.py
This patch adds support for the “--submit” parameter in the gnt-instance
script, for the commands where it makes sense.
---
scripts/gnt-instance | 46 +-
1 files changed, 29 insertions(+), 17 deletions(-)
diff --git a/scripts/gnt-instance
On Thu, Sep 11, 2008 at 07:40:28AM -0700, Alexander Schreiber wrote:
On Thu, Sep 11, 2008 at 7:29 AM, Guido Trotter [EMAIL PROTECTED] wrote:
On Thu, Sep 11, 2008 at 3:27 PM, Alexander Schreiber [EMAIL PROTECTED]
wrote:
Hello ganeti-devel@googlegroups.com,
I'd like you to do a code
On Thu, Sep 11, 2008 at 08:18:16AM -0700, Alexander Schreiber wrote:
Hello ganeti-devel@googlegroups.com,
I'd like you to do a code review on the patch below:
Make burnin aware of hvm device type flags.
Merged from branches/ganeti/ganeti-1.2 r1648
Use static values for new hvm
On Thu, Sep 11, 2008 at 08:42:31AM -0700, Alexander Schreiber wrote:
Hello ganeti-devel@googlegroups.com,
I'd like you to do a code review on the patch below:
Make burnin aware of hvm device type flags.
Merged from branches/ganeti/ganeti-1.2 r1648
Use static values for new hvm
On Thu, Sep 11, 2008 at 05:15:11PM +0100, Guido Trotter wrote:
It used to refer to nodes, which was confusing.
---
scripts/gnt-instance |6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)
wierd... LGTM!
iustin
On Wed, Sep 17, 2008 at 05:34:53PM +0200, Michael Hanselmann wrote:
It abstracts exception handling and is like a complement to
utils.WriteFile.
---
lib/utils.py | 17 +
1 files changed, 17 insertions(+), 0 deletions(-)
diff --git a/lib/utils.py b/lib/utils.py
index
On Thu, Sep 18, 2008 at 10:31:27AM +0100, Guido Trotter wrote:
There is no such feature in trunk yet.
---
lib/cmdlib.py |1 -
tools/burnin |2 --
2 files changed, 0 insertions(+), 3 deletions(-)
LGTM.
On Thu, Sep 18, 2008 at 10:31:28AM +0100, Guido Trotter wrote:
Currently when not locking all nodes/instances are returned, regardless
if the user asked only for some of them. (This breaks burnin)
---
lib/cmdlib.py |4
1 files changed, 4 insertions(+), 0 deletions(-)
But since
On Thu, Sep 18, 2008 at 11:32:42AM +0100, Guido Trotter wrote:
Currently when not locking all nodes/instances are returned, regardless
if the user asked only for some of them. With this patch we return to
the previous behaviour:
- if no names are specified return info on all current ones
The code in 'updating instance configuration' section of the replace
disks with change secondary node was setting a wrong new logical_id for
the drbd devices (only set the new node, not the new minor). The patch
fixes this by remembering the new logical_id that we use for activating
the drbd and
This patch adds support for allocating static minors.
Like for the LVM
uuids, we add a new cache for the temporarily allocated requests, and
the users of the new methods must manually clear the cache. If this
doesn't happen, at worst we lose some minors.
Todos remaining:
- implement duplicate
On Tue, Sep 23, 2008 at 11:33:17AM +0100, Guido Trotter wrote:
---
configure.ac |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/configure.ac b/configure.ac
index c6d6416..e1f4f76 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1,6 +1,6 @@
# Configure
With some todos remaining, this patch switches the DRBD devices to use
the passed minors, and the cmdlib code (add instance and replace disks)
to request and assign minors to the DRBD disks.
Todos:
- look at the disk RPC calls to see which can be optimized away, since
we now know the minor
A couple of more modules are using the obsolete logger functions, config
being one of them.
---
lib/config.py |7 +++
1 files changed, 3 insertions(+), 4 deletions(-)
diff --git a/lib/config.py b/lib/config.py
index 8198b1c..e24a17d 100644
--- a/lib/config.py
+++ b/lib/config.py
@@
On Wed, Sep 24, 2008 at 03:43:33PM +0200, Michael Hanselmann wrote:
Future patches will add even more variables to the cluster config.
Adding more parameters wouldn't make the function easier to use and
it doesn't make sense to pass them to another function, as it's
only done once in
It's not complete, but I hope it's up to date. It's restructured text,
but no make rules or such for html output are needed, it's readable
as-is.
---
doc/design-2.0-master-daemon.rst | 190 ++
1 files changed, 190 insertions(+), 0 deletions(-)
create mode
---
doc/design-2.0-disk-handling.rst | 259 ++
1 files changed, 259 insertions(+), 0 deletions(-)
create mode 100644 doc/design-2.0-disk-handling.rst
diff --git a/doc/design-2.0-disk-handling.rst b/doc/design-2.0-disk-handling.rst
new file mode 100644
index
On Wed, Sep 24, 2008 at 03:43:35PM +0200, Michael Hanselmann wrote:
This will be used to read the configuration file in the node daemon.
The write functionality is needed for master failover.
---
Makefile.am |1 +
lib/simpleconfig.py | 92
On Wed, Sep 24, 2008 at 03:43:39PM +0200, Michael Hanselmann wrote:
The configuration version is now again in the configuration file.
---
lib/config.py | 22 --
1 files changed, 8 insertions(+), 14 deletions(-)
Looks good, but maybe to have this check outside of the
On Wed, Sep 24, 2008 at 03:43:40PM +0200, Michael Hanselmann wrote:
Replacing ssconf with configuration.
---
lib/mcpu.py | 10 --
test/mocks.py |6 ++
2 files changed, 10 insertions(+), 6 deletions(-)
LGTM, thanks.
iustin
On Wed, Sep 24, 2008 at 03:43:43PM +0200, Michael Hanselmann wrote:
Get rid of ssconf and convert to configuration instead.
---
Hmm... Patch LGTM, but I have more doubts now that I realize that for
each ssh connection and such we open and parse the whole config file...
I did a very quick
On Wed, Sep 24, 2008 at 03:43:44PM +0200, Michael Hanselmann wrote:
Replacing ssconf with configuration. Cluster rename is broken and stays
that way.
---
lib/cmdlib.py | 96
-
1 files changed, 47 insertions(+), 49 deletions(-)
On Wed, Sep 24, 2008 at 03:43:45PM +0200, Michael Hanselmann wrote:
Replace ssconf with configuration.
---
lib/bootstrap.py | 31 +++
1 files changed, 15 insertions(+), 16 deletions(-)
LGTM, thanks.
iustin
This patch adds the ‘serial_no’ attribute to the other top-level objects
(the configuration object itself, the nodes and the instances).
---
lib/objects.py |4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)
diff --git a/lib/objects.py b/lib/objects.py
index a26907d..23ecfc8 100644
This patch add initialization of the serial_no on instance and nodes,
and update of the field whenever an object is updated in the generic
case, via ConfigWriter.Update(obj) and in the specific case of
instances' state being modified manually.
---
lib/config.py |7 ++-
1 files changed, 6
This seems to be the last (deprecated) use of AddInstance in order to
update an instance.
---
lib/cmdlib.py |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/cmdlib.py b/lib/cmdlib.py
index 378aab5..6a97526 100644
--- a/lib/cmdlib.py
+++ b/lib/cmdlib.py
@@ -2373,7
This patch adds listing of the serial_no attribute in gnt-instance and
gnt-node list, and updates to the manpages to reflect the change.
---
lib/cmdlib.py |6 ++
man/gnt-instance.sgml |9 +
man/gnt-node.sgml |9 +
scripts/gnt-instance |7 +--
On Sat, Sep 27, 2008 at 02:08:55PM +0100, Guido Trotter wrote:
On Sat, Sep 27, 2008 at 11:37 AM, Iustin Pop [EMAIL PROTECTED] wrote:
This seems to be the last (deprecated) use of AddInstance in order to
update an instance.
LGTM
---
lib/cmdlib.py |4 ++--
1 files changed, 2
On Sat, Sep 27, 2008 at 02:22:21PM +0100, Guido Trotter wrote:
On Sat, Sep 27, 2008 at 11:37 AM, Iustin Pop [EMAIL PROTECTED] wrote:
This patch adds update of the cluster serial number for:
- add/remove node (as the cluster's node list is changed)
- add/remove/rename instance
On Sat, Sep 27, 2008 at 07:13:38PM +0100, Guido Trotter wrote:
On Sat, Sep 27, 2008 at 1:04 AM, Iustin Pop [EMAIL PROTECTED] wrote:
In case the config file is manually modified, or in case of bugs, the
tcp/udp ports could be reused, which will create various problems
(instances not able
and are not deemed as sensitive), but instead to
prevent mistakes (especially with the new static minors changes, which
are not yet proved stable).
Iustin Pop (4):
Move the pseudo-secret generation to utils.py
Extend DRBD disks with shared secret attribute
Enable the use of shared secrets
Move
For now we only use the ‘C’ protocol so we can put it in constants.py
instead of hardcoding it.
---
lib/bdev.py |9 ++---
lib/constants.py |1 +
2 files changed, 7 insertions(+), 3 deletions(-)
diff --git a/lib/bdev.py b/lib/bdev.py
index 8910f80..2d1 100644
---
This patch, which is similar to r1679 (Extend DRBD disks with minors
attribute), extends the logical and physical id of the DRBD disks with a
shared secret attribute. This is generated at disk creation time and
saved in the config file.
The generation of the secret is done so that we don't have
When the node does not yet have any minors allocated, the first minor
(0) will not be entered in the ConfigWriter._temporary_drbds structure.
This does not happen for our current usage, since we always ask for two
minors (so the next call will not match this case), but it will be
triggered if we
Currently, it is hard to examine a job in detail; the output of ‘gnt-job
list’ is not easy to parse.
The patch adds a ‘gnt-job info’ command that is (vaguely) similar to
‘gnt-instance info’ in that it shows in a somewhat easy to understand
format the details of a job.
The result formatter is
Unless we decide to change the job identifiers to integer, we should at
least sort the list returned by _GetJobIDsUnlocked.
---
lib/jqueue.py |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/lib/jqueue.py b/lib/jqueue.py
index e55f46d..b44b863 100644
--- a/lib/jqueue.py
after people
are fine with the commands as proposed, otherwise I'm afraid it might be
too early (since they might change).
Iustin Pop (3):
Nicely sort the job list
Implement job summary in gnt-job list
Add a info subcommand to gnt-job
lib/jqueue.py |4 ++-
lib/opcodes.py | 34
It is not currently possibly to show a summary of the job in the output
of “gnt-job list”. The closes is listing the whole opcode(s), but that
is too verbose. Also, the default output (id, status) is not very
useful, unless one looks for (and knows about) an exact job ID.
The patch adds a
This patch adds the job execution log in “gnt-job info” and also allows
its selection in “gnt-job list” (however here it's not very useful as
it's not easy to parse). It does this by adding a new field in the query
job call, named ‘oplog’.
With this, one can get a very clear examination of the
On Tue, Sep 30, 2008 at 12:03:02PM +0200, Michael Hanselmann wrote:
2008/9/30 Iustin Pop [EMAIL PROTECTED]:
This patch adds start, stop, and received timestamp for jobs (and allows
querying of them), and allows querying of the opcode timestamps.
---
Not very sure about where we update
On Tue, Sep 30, 2008 at 02:05:32PM +0200, Iustin Pop wrote:
This patch adds posibility of selection of job/opcode timestamps in
gnt-job list and info.
The code handling the possible cases (None or a valid timestamps) are
ugly though...
---
scripts/gnt-job | 65
On Tue, Sep 30, 2008 at 02:17:25PM +0200, Michael Hanselmann wrote:
2008/9/30 Iustin Pop [EMAIL PROTECTED]:
On Tue, Sep 30, 2008 at 02:13:50PM +0200, Michael Hanselmann wrote:
This patch also adds the design documents to Makefile.am.
---
Makefile.am | 13
On Tue, Sep 30, 2008 at 02:28:46PM +0200, Michael Hanselmann wrote:
---
DEVNOTES | 10 ++
1 files changed, 10 insertions(+), 0 deletions(-)
LGTM, thanks.
iustin
Curently PollJob accepts a generic job, but will return (history
artifact) only the first opcode result. This is wrong, as it doesn't
allow polling of a job with multiple results.
Its only caller (for now) is also changed, so no functional changes
should happen.
---
lib/cli.py |6 --
1
On Wed, Oct 01, 2008 at 10:59:35AM +0200, Iustin Pop wrote:
The watcher was using conflicting attributes of the instance:
- it queried the admin_/oper_state, which are booleans
- but it compared those to the status (which is a text field)
The code was changed to query only
The watcher was using conflicting attributes of the instance:
- it queried the admin_/oper_state, which are booleans
- but it compared those to the status (which is a text field)
The code was changed to query the aggregated 'status' field, as that
will also return indication of node
On Fri, Oct 03, 2008 at 11:59:16AM +0100, Guido Trotter wrote:
On Thu, Oct 2, 2008 at 10:38 AM, Iustin Pop [EMAIL PROTECTED] wrote:
The patch changed the xen hypervisor to compute the number of cpu
sockets/nodes and enables the command line and the RAPI to show this
information (for RAPI
In case the job object doesn't have a timestamp (which is a separate
issue), the listing should not break. We fix this by changing the
FormatTimstamp function itself to return '?' in case the timestamp
doesn't look good (note that it still can break if non-integers are
returned, but this is
Since our locks are not gathered nicely, we can have jobs that are
actually blocking on locks (parallel burnin shows this), so at least we
need to increase the number of threads above the usual number of jobs we
could have in a such a case.
---
lib/jqueue.py |2 +-
1 files changed, 1
This patch introduces a simple framework for executing jobs in parallel
in burnin (the ExecJobSet function) and the --parallel command line
flag.
The patch also changes the instance creation to run in parallel when the
above flag is given. Error handling/instance removal is currently flacky
with
On Mon, Oct 06, 2008 at 03:34:57PM +0200, Iustin Pop wrote:
This patch introduces a simple framework for executing jobs in parallel
in burnin (the ExecJobSet function) and the --parallel command line
flag.
The patch also changes the instance creation to run in parallel when the
above flag
Currently there are three function in backend that need the cluster name
in order to instantiate an SshRunner. The patch changes these to get the
cluster name from the master in the rpc call; once the multi-hypervisor
change is implemented, then very few places in which we need the SCR
remain in
1 - 100 of 8206 matches
Mail list logo
