workerpool: Make worker ID alphanumeric
Having a proper name instead of just a number makes debuggingeasier.
Signed-off-by: Michael Hanselmann <hansmi@google.com>Reviewed-by: Iustin Pop <iustin@google.com>
locking: Fix race condition in LockSet
This patch fixes a race condition when acquiring all locks ina LockSet instance. The list of lock names needs to be sortedto guarantee a consistent locking order, but the names were notsorted when acquiring all locks in the set....
mcpu: Log lock status with sorted names
Reading and comparing sorted lists is easier when debugging locking problems.
locking: Append to list outside error handling block
Signed-off-by: Michael Hanselmann <hansmi@google.com>Reviewed-by: Guido Trotter <ultrotter@google.com>
locking: Don't fail in error handling if lock isn't owned
In case an exception was thrown while acquiring the lock, not necessarily allowned locks are also really acquired. Before this change, an exception could bemasked by another exception thrown here. There is no good clean-up strategy...
Normalize MAC addresses to all lower.
This change will normalize the MAC to all lower after validation.
Signed-off-by: René Nussbaumer <rn@google.com>Signed-off-by: Iustin Pop <iustin@google.com>Reviewed-by: Iustin Pop <iustin@google.com>
Introduce a Luxi call for GetTags
This changes from submitting jobs to get the tags (in cli scripts) toqueries, which (since the tags query is a cheap one) should be muchfaster.
The tags queries are already done without locks (in the generic querypaths for instances/nodes/cluster), so this shouldn't break tags query...
LURenameCluster: run post hook on all nodes
Since the cluster name might be used for various purposes on nodes, weshould let all nodes "know" about a cluster rename by running the posthook on all nodes. This will make cluster rename slightlyslower/costlier, but it is not/shouldn't be an operation that is run...
Fix unused imports or add silences where needed
In some cases pylint doesn't parse the import correctly, so we addsilences; but there are also many cases of unused imports, which wesimply remove.
Signed-off-by: Iustin Pop <iustin@google.com>Reviewed-by: Olivier Tharan <olive@google.com>
bdev: Add a TODO and a pylint silence
A piece of old code in bdev.py uses a for loop over a single variablebecause we can 'break' out of the loop or exit on the 'else' path. Thisis not a nice usage of the for loop, it should be converted to astandard if...elif...else structure....
Further pylint disables, mostly for Unused args
Many of our functions have to follow a given API, and thus we have tokeep a given signature, but pylint doesn't understand this. Therefore,we silence this warning.
The patch does a few other cleanups.
Signed-off-by: Iustin Pop <iustin@google.com>...
LUDiagnoseOS._DiagnoseByOS: remove unused arg
The node_list argument to _DiagnoseByOS is not used, and is obsoleted bythe fact that the rlist argument already has the valid nodes as keys(assuming RPC behaviour didn't change). Thus, we remove it and silence...
hv_xen/_GetConfigFileDiskData: remove unused arg
The disk template is not needed, all that's used is the disk data. Assuch, remove this parameter from the function.
jqueue/_CheckRpcResult: log the whole operation
Currently only the rpc call, but not its description (which also showsthe argument) is logged. We change this to log failmsg too, and thisalso silences a warning.
Optparse extenders have to obey a given API
So we just silence the warning.
backend._OSOndiskAPIVersion: remove obsolete arg
The 'name' argument is not used anymore, probably since before 2.0.Since this is an internal function, we can just remove it (from itscaller too).
pylint cleanups: dangerous initializers
Plus a silence for a wrong "uninitialized var".
Convert to static methods (where appropriate)
Many methods are simple pure functions, and not depending on the objectstate. We convert these to staticmethods.
Add targeted pylint disables
This patch should have only:
- pylint disables- docstring changes- whitespace changes
Implement all hv functions in hv_chroot/hv_fake
The chroot and fake hypervisors were missing:
- the powercycle node functionality- proper handling of migration requests
The powercycle was just used as in the other hypervisors (use thestandard linux powercycle). The migration for chroot was disabled...
Add some stubs to bdev.FileStorage
This patch adds explicit errors (instead of notimplemented) inFileStorage (and the associated TODOs).
Fix indentation issues
Fix two bugs in ConfigWriter._EnsureUUID
Wrong argument name and wrong number of arguments in string formatting.
KVM: Abstract/rework instance up checks
This patch abstract the check "is instance stopped" into a separatefunction, and thus simplifies a couple of higher-level functions. Italso moves from manual read of the pidfile to use the (correctabstraction of) _InstancePidAlive....
KVM: Split out the pidfile computation
In some cases we only need the pidfile, but not the pid or the alivestatus.
Fix an error message
Detected by an 'Unused variable' warning.
Remove many 'Unused variable' warnings
Note there are some cases left which need extra cleanup.
Fix use of the logging functions
The logging functions expand the arguments themselves, thus it's saferto let them do it rather than manual string formatting.
Also re-wraps one comment.
Merge branch 'devel-2.0' into devel-2.1
Add targetted pylint disables
This patch adds targeted pylint disables, where it makes sense (eitherdue to limitations in pylint or due to historical usage), and also a fewblanket ones in rapi where all the names are… “different”.
Fix two bugs in seldom-used codepaths
New version of pylint, new bugs found!
Clarifiy some more wide pylint disables
This removes/updates some module-wide pylint disables.
Implement BuildHooksEnv for NoHooksLU
This just adds a stub function that raises an assertion error; thisaccomplishes two things:
- silences many pylint warnings- if we ever stumble upon this, a specific assertion error is (hopefully) clearer than just a not implemented error...
Fix indentation in hv_kvm
Per pylint warnings.
Partial cherry-pick of 6c881c5 from the 2.1 branch
This cherry-picks the utils.FieldSet.Matches changes and the significantjqueue.py change. These are stable in the 2.1 branch and therefore makesense to backport to 2.0 (are basically cleanups).
Fix a typo in the doc string
MaybeRaise in lib/errors.py had a typo in the doc string
Merge branch 'stable-2.1' into devel-2.1
Merge branch 'stable-2.0' into stable-2.1
CreateInstance: allow no ip check with start mode
Since gnt-instance start doesn't do any checks on the IP, it doesn'tmake much sense to do so in instance create (with start) if the userexpressly passes in ‘--no-ip-check’. Removing this requirement eases the...
Command line/RAPI support for --no-name-check
This patch adds --no-name-check to gnt-instance add and gnt-backupimport. This is opposite to the opcode parameter (name_check) as it issimilar to ip_check and start.
It also adds it to RAPI and gnt-instance batch-create as a parameter in...
Op/LUCreateInstance support for (no) name checks
This adds a new opcode parameter ‘name_check’ (similar to ip_check) thatis not required to be present (to easy backwards compatibility fortools).
It also adds a CheckArguments to LUCreateInstance and changes the...
Pass --fqdn to ssh hostname checks
The cluster verify checks for fqdn are done via address lookups, andthere we actually use the FQDN. However, for the ssh hostname checkwhich is done at node add time, we rely on the default of the “hostname”command. And Debian for example recently changed the default to return...
Move the hooks file mask into constants.py
This will allow reuse of the same mask for multiple validations.
Signed-off-by: Iustin Pop <iustin@google.com>Reviewed-by: Michael Hanselmann <hansmi@google.com>
Security issue: add validation of script names
This patch unifies the search for external script to always go throughutils.FindFile and implements in that function a restriction on validchars in file names and (additionally) that the passed name is the...
Improve LUQueryNodes for lockless case
In most uses of LUQueryNodes, we don't take a lock. This means that theinstance data is not protected across GetInstanceList andGetInstanceInfo, and this can lead to instances not existing anymore.
Switching to GetAllInstanceInfo means that we get a single,...
Add disk cache control parameter for KVM
This patch adds the 'cache' parameter for KVM; currently this is onlycustomisable at the hypervisor level, so it's the same for all drives(except any CDROM image, which gets the default).
Change pyinotify import for broader compatibility
On some distributions pyinotify is installed in a different way, and theactual module just contains an internal pyinotify entry, which is theactual library. On others the main pyinotify module contains the library...
ClusterMasterQuery: add primary ip field
By allowing also the primary ip field to be fetched directly, we avoidone more confd lookup, or dns request, to find out which address themaster node lives at.
Signed-off-by: Guido Trotter <ultrotter@google.com>
confd ClusterMasterQuery: allow fields request
Change the ClusterMasterQuery to allow a query, and if present accept alist of fields to be returned. Currently only name and ip are accepted.
This feature will be used by NLD to route the cluster ip over the nbma....
DRBD: ignore unreadable meta devices
The DRBD driver can ignore dead disks but not dead meta devices (forwhich it refuses to configure the minor). To handle this case, we checkthat the meta device is readable and if not we ignore it (the same aswhen backend._RecursiveAssembleBD didn't find it)....
Simplify utils.ReadFile
The documentation for file objects' read method states that if the sizeis "negative or ommitted", all data is read; thus we can simplify it tohave size=-1 by default and not have the if test.
gnt-cluster verify: Warn if node time diverges too far
The warning will be generated if the clocks diverge by morethan 150 seconds. Due to the way the RPC system works, wecannot get exact time differences, e.g. if one of thequeried nodes is broken. The comparision is done using a...
KVM: fail when a routed nic has no ip
This shouldn't happen, but if it does it's better to fail at this level,rather than create a broken NIC script, which is hard to debug.
Signed-off-by: Guido Trotter <ultrotter@google.com>Reviewed-by: Iustin Pop <iustin@google.com>
cmdlib: Work around race condition in DRBD before version 8.0.13
DRBD goes into sync mode for a short amount of time afterexecuting the "resize" command. DRBD 8.x below version8.0.13 contains a bug whereby calling "resize" in syncmode fails.
Signed-off-by: Michael Hanselmann <hansmi@google.com>...
Remove quotes from CommaJoin and convert to it
This patch removes the quotes from CommaJoin and converts most of thecallers (that I could find) to it. Since CommaJoin does str(i) for i inparam, we can remove these, thus simplifying slightly a few calls....
Re-add “nic.bridges” field to RAPI bulk instance list
Commit 495cfdf0 removed “nic.bridges” from the defaultlist for bulk instance list RAPI requests.
Revert "Get rid of utils.CommaJoin"
This reverts commit 6915bc28fe053e92aa16cf2d974d205f1140219c based on thread onganeti-devel.
Conflicts:
lib/cmdlib.py (due to the error code classification, trivial)
Add check for OpenSSL entropy status
By checking for this explicitly, the errors (SSLEAY_RAND_BYTES, “PRNGnot seeded”) will happen in the start-up phase of the daemon and notonly when executing remote procedure calls.
Handle EEXIST in utils.RenameFile
This should fix an issue I've seen exactly once during testing. It might havebeen caused by parallel RPC calls to archive jobs.
[…] ganeti-noded:112 ERROR Error in RPC call […] File "/usr/lib/python2.4/site-packages/ganeti/backend.py", line 2365, in JobQueueRename...
Remove unused parameter “unlock” from cmdlib._WaitForSync
Fix off-by-one error when modifying instance NIC
For an instance with exactly one NIC:
$ gnt-instance modify --net 1:ip=1.2.3.4 inst1Failure: prerequisites not met for this operation:error type: wrong_input, error details:Invalid NIC index 1, valid values are 0 to 1...
Re-add check for duplicate instance IP
This was originally implemented in 0ce8f948 and partiallyrolled back in 9b65e0d4. Apart from re-adding the check,this patch does some housekeeping by renaming the “_helper”function to “_AddIpAddress”.
config: Style fixes
Fix epydoc error
Fix change of cluster nic parameters
To stay on the safe side, we check for errors in all instances, andrefuse to act, reporting on the errors we found, if there are anyproblems.
Signed-off-by: Guido Trotter <ultrotter@google.com>Reviewed-by: Michael Hanselmann <hansmi@google.com>
Fix mispopulation of nic parameters at nic modify
There's a bug in Ganeti 2.1 rc0 that makes nic parameters be populatedfrom the "filled in" dict, even if we're not changing any values inthem. This patch fixes the problem, by populating them from the correct...
NIC.CheckParameterSyntax: fix bridged check
We should match for the strings to be the same "==" not to point to thesame memory location with is, or we skip the actual check.
Fix and simplify socat escape detection
- Program paths should not be --with-… options (see Autoconf docs)- Simplify checks for escape functionality- Make SOCAT_USE_ESCAPE variable a bool
Use “daemon-util” to reload SSH keys
KVMHypervisor: fix broken error format string
ConfigWriter: move _temporary_ids to reservation
In order to do this we need to pass a job id when reserving a resource.We have one during _EnsureUUIDs because we passed it in from AddNode andAddInstance. During config upgrade we use a fake job ID which we then...
ConfigWriter: move _temporary_macs to reservation
This solves the race conditions in mac reservation, as macs are actuallyreserved, under the current ec id.
ConfigWriter: simplify GenerateDRBDSecret
We can do this by adding a new TemporaryReservationManager
TemporaryReservationManager
Add errors.ReservationError
Remove exceptions list from GenerateUniqueID
It's not used anywhere, so it's dead code.
Processor: support a unique execution id
When the processor is executing a job, it can export the execution id toits callers. This is not supported for Queries, as they're not executedin a job.
Add config.DropECReservations
For now this function does nothing, but it gets called by mcpu when theexecution of an LU is done, making sure any pending reservations aredropped.
config.Add{Node,Instance}: get the ec id
This is ok because adding a node or instance cannot happen in a query.
We get the ec id from the LU and pass it to _EnsureUUID, which willthen for now not use it.
Signed-off-by: Guido Trotter <ultrotter@google.com>...
Fix pylint 'E' (error) codes
This patch adds some silences and tweaks the code slightly so that“pylint --rcfile pylintrc -e ganeti” doesn't give any errors.
The biggest change is in jqueue.py, the move of _RequireOpenQueue out ofthe JobQueue class. Since that is actually a function and not a method...
Add use_localtime parameter for xen-hvm and kvm
Currently xen-hvm and kvm use different real time clock by default. Toreduce confusion, this patch adds an optional use_localtime parameter.
If the real time clock on the instance is set to local time, the...
Introduce 'global hypervisor parameters' support
This patch adds support for global hypervisor parameters in instancecreation, instance modification, instance query and at instance loadtime.
We basically prevent any query on these parameters, discard them at load...
Remove the KVM_MIGRATION_PORT configure.ac param
Since this is easily configurable at run-time, we remove theconfigure-time parameter. If anyone is building custom packages, thenthe default can be tweaked by a one-line patch to constants.py.
Note that this also fixes the type of parameter, the default from...
gnt-*: Print better error message for uninitialized cluster
Cache JSON encoders and sort keys
The sort_keys argument is supported since simplejson 1.3.
Add new “daemon-util” script to start/stop Ganeti daemons
Until now, Ganeti started and stopped its own daemons using custom functions.To start, the daemon was just executed and then sent the appropriate signals tostop it again. Init scripts would have to pay attention to the PID file and...
kvm console: use socat raw mode with escape
If this is enabled at configure time, we pass in different parameters tothe socat console, making it a lot more manageable.
Migration: add check for listening target
This patch adds a check for listening on the remote port in Xen and KVMmigrations. This will be generating a single “load of migration failed”message for KVM, but otherwise not prevent the migration. For Xen (which...
TLMigrateInstance: add error messagess during Exec
Currently the migration of an instance doesn't show any error until theend. We add two messages that show better the progress:
node1# gnt-instance migrate -f instance5Wed Nov 4 04:04:34 2009 Migrating instance instance5...
hypervisors: switch to using HV_MIGRATION_PORT
This changes KVM to use HV_MIGRATION_PORT instead of KVM_MIGRATION_PORTand enables passing the port for Xen migrations.
Since KVM_MIGRATION_PORT is not used anymore, we stop exporting it fromconstants.py....
Introduce HV_MIGRATION_PORT hypervisor parameter
This parameter will replace the direct use of KVM_MIGRATION_PORT and theimplicit use of the Xen migration port.
While it doesn't make sense to change this at instance level, we don'thave any other infrastructure for cluster-wide hypervisor parameters, so...
hypervisors: change MigrateInstance API
Currently the $hypervisor.MigrateInstance takes the instance name. Thispatch changes it to take the instance object, such that other instanceproperties (especially hvparams) are available to it.
Revert the instance IP conflicts
Since instances can live in different VLANs from nodes (especially inrouted mode), based on the 'link' parameter, we shouldn't alwaysrestrict having duplicate IPs. Thus we only check the node IPs/clusterIP for now.
Introduce a wrapper for hostname resolving
Currently a few of the LU's CheckPrereq use utils.HostInfo which raisesa resolver error in case of failure. This is an exception from thestandard that CheckPrereq should raise an OpPrereqError if the error isin the 'pre' phase (so that it can be retried)....
Add a configuration verify check for duplicate IPs
This patch adds a check that the cluster IP, the nodes primary (andsecondary, if enabled) IP and the instances NIC IPs are unique in thecluster.
Workaround fake failures in drbd+live migration
This patch is an attempt to fix the ugly issue during migration: Cannot resync disks on node …: [True, 100]
If my understanding is correct, sometimes we poll the /proc/drbd file atan inoportune moment, while it's being updated, or while the DRBD device...
Another round of pylint-related style fixes
A newer version of pylint, more warnings…
Revert "kvm console: use socat raw mode with escape"
This reverts commit ce0eb6694e3fb2510035501539c7acc92a0f174e, since it dependson 37fc2cf5ba8919cef407199ee540aad4b1a9a2b6 which will be reverted too.
Change behaviour of ConfigWriter._WriteConfig
This patch changes the behaviour of _WriteConfig in case ofconfiguration errors:
- before, it used to abort the saving (even though the in-memory configuration used by current jobs has already changed)- now, we log it (both to the log and to the user) but continue, since...