It takes a variable amount of time for the multipath daemon
to create /dev/dm-XX in response to new LUNs being discovered.
The old iscsi_util code only discovered the multipath device
if it was created quickly enough, but in a significant number
of cases, kubelet would grab one of the individual paths and
put a filesystem it on before multipathd could construct a
multipath device.
This change waits for the multipath device to get created for
up to 10 seconds, but only if the PV actually had more than
one portal.
This change ensures that iSCSI block devices are deleted after
unmounting, and implements scanning of individual LUNs rather
than scanning the whole iSCSI bus.
In cases where an iSCSI bus is in use by more than one attachment,
detaching used to leave behind phantom block devices, which could
cause I/O errors, long timeouts, or even corruption in the case
when the underlying LUN number was recycled. This change makes
sure to flush references to the block devices after unmounting.
The original iSCSI code scanned the whole target every time a LUN
was attached. On storage controllers that export multiple LUNs on
the same target IQN, this led to a situation where nodes would
see SCSI disks that they weren't supposed to -- possibly dozens or
hundreds of extra SCSI disks. This caused 3 significant problems:
1) The large number of disks wasted resources on the node and
caused a minor drag on performance.
2) The scanning of all the devices caused a huge number of uevents
from the kernel, causing udev to bog down for multiple minutes in
some cases, triggering timeouts and other transient failures.
3) Because Kubernetes was not tracking all the "extra" LUNs that
got discovered, they would not get cleaned up until the last LUN
on a particular target was detached, causing a logout. This led
to significant complications:
In the time window between when a LUN was unintentially scanned,
and when it was removed due to a logout, if it was deleted on the
backend, a phantom reference remained on the node. In the best
case, the phantom LUN would cause I/O errors and timeouts in the
udev system. In the worst case, the backend could reuse the LUN
number for a new volume, and if that new volume were to be
scheduled to a pod with a phantom reference to the old LUN by the
same number, the initiator could get confused and possibly corrupt
data on that volume.
To avoid these problems, the new implementation only scans for
the specific LUN number it expects to see. It's worth noting that
the default behavior of iscsiadm is to automatically scan the
whole bus on login. That behavior can be disabled by setting
node.session.scan = manual
in iscsid.conf, and for the reasons mentioned above, it is
strongly recommended to set that option. This change still works
regardless of the setting in iscsid.conf, and while automatic
scanning will cause some problems, this change doesn't make the
problems any worse, and can make things better in some cases.
These are all flagged by Go 1.11's
more accurate printf checking in go vet,
which runs as part of go test.
Lubomir I. Ivanov <neolit123@gmail.com>
applied ammend for:
pkg/cloudprovider/provivers/vsphere/nodemanager.go
Google's configure-helper.sh script bind-mounts /var/lib/kubelet somewhere
into /home/kubernetes and thus every mount that Kubernetes does is visible
twice in /proc/mounts.
iSCSI and RBD should not rely on counting on entries in /proc/mounts and
unmount device when Kubernetes thinks it's unusued. Kubernetes tracks
the mounts by itself and most of other volume plugins rely on it safely.
This PR add comments for the background why plugin gets loopback
device and removes loopback device even if operation_generator has
same functionality.
Automatic merge from submit-queue (batch tested with PRs 57399, 57751, 57475). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
iscsi: set node.startup to manual
If the default iSCSI node.startup is set to automatic, if there is a node failure,
any pods on that node will get rescheduled to another node. If the failed node is
later brought back up it will then try to log back in to any iSCSI sessions it had
prior to the failure, which may no longer exist or may be now in-use by the other
nodes.
It appears most platforms keep the open-iscsi default of node.startup-automatic.
But in case this system-wide setting has been changed, and just to be explicit, this
sets node.startup values for kubernetes controlled volumes to manual.
Fixes#21305
```release-note
iSCSI sessions managed by kubernetes will now explicitly set startup.mode to 'manual' to
prevent automatic login after node failure recovery. This is the default open-iscsi mode, so
this change will only impact users who have changed their startup.mode to be 'automatic'
in /etc/iscsi/iscsid.conf.
```
If the default iSCSI node.startup is set to automatic, if there is a node failure,
any pods on that node will get rescheduled to another node. If the failed node is
later brought back up it will then try to log back in to any iSCSI sessions it had
prior to the failure, which may no longer exist or may be now in-use by the other
nodes.
It appears most platforms keep the open-iscsi default of node.startup-automatic.
But in case this system-wide setting has been changed, and just to be explicit, this
sets node.startup values for kubernetes controlled volumes to manual.
Closes issue #21305
Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Rename Detach() parameter.
`Detach()` does not get device name, it gets volume name. Parameters named `deviceMountPath` or `deviceName` just confuses developers.
Note that this PR just renames parameters here and there, there should be no behavior change.
@kubernetes/sig-storage-pr-reviews
/assign @gnufied @jingxu97
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.
Update volume OWNERS to reflect active sig-storage reviewers
**What this PR does / why we need it**:
Update sig-storage reviewers to add new members and remove those that don't have as much time to review storage PRs. Approvers are unchanged.
**Special notes for your reviewer**:
For all those that have been removed, please approve. If you want to remain as a reviewer, let me know and I will add you back.
**Release note**:
NONE
This PR makes following changes.
- Simplify volume tearDown path for iSCSI and FC using
util.UnmountPath().
- Log lastErr during iscsi connection
If iscsid fails to connect second portal, currently
the error is ignored silently. The lastErr should be
logged to find the root cause of problem.
- Remove iscsi plugin directory after iscsi connection
is successfully closed.
Automatic merge from submit-queue (batch tested with PRs 51628, 51637, 51490, 51279, 51302)
Fix iSCSI WaitForAttach not mounting a volume
WaitForAttach failed consistently with this error:
```
Heuristic determination of mount point failed:stat /var/lib/kubelet/plugins/kubernetes.io/iscsi/iface-default/10.128.0.3:3260-iqn.2003-01.org.linux-iscsi.f21.x8664:sn.4b0aae584f7c-lun-0: no such file or directory
```
We should ignore "no such file or directory" eror, the directory is created just few lines below.
Fixes: #51625
**Release note**:
```release-note
NONE
```
/sig storage
/assign @mtanino
WaitForAttach failed consistently with this error:
Heuristic determination of mount point failed:stat /var/lib/kubelet/plugins/kubernetes.io/iscsi/iface-default/10.128.0.3:3260-iqn.2003-01.org.linux-iscsi.f21.x8664:sn.4b0aae584f7c-lun-0: no such file or directory
We should ignore "no such file or directory" eror, the directory is created
just few lines below.
Automatic merge from submit-queue (batch tested with PRs 51439, 51361, 51140, 51539, 51585)
Iscsi handle failed mount
**What this PR does / why we need it**:
**Which issue this PR fixes**: fixes#50556
**Special notes for your reviewer**:
**Release note**:
```release-note
None
```
Automatic merge from submit-queue
Fix swallowed error in iscsi package
**What this PR does / why we need it**: Fixes a swallowed error in the iscsi package.
```release-note NONE
```
Automatic merge from submit-queue (batch tested with PRs 51108, 51035, 50539, 51160, 50947)
iscsi: Use VolumeHost.GetExec() to execute stuff in volume plugins
**What this PR does / why we need it**:
This PR updates iSCSI volume plugin to use `VolumeHost.GetExec()` to execute utilities like mkfs and lsblk instead of simple `os/exec`. This prepares the volume plugin to run these utilities in containers instead of running them on the host + makes the volume plugin more independent and less hardcoded.
See proposal in https://github.com/kubernetes/community/pull/589.
Note that this PR does **not** change place where the utilities are executed - `VolumeHost.GetExec()` still leads directly to `os/exec`. It will be changed when the aforementioned proposal is merged and implemented.
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue (batch tested with PRs 38947, 50239, 51115, 51094, 51116)
Call Fatalf instead of Errorf when mounter/unmounter is nil in volume tests
If we get a nil mounter or unmounter, a panic will occur. Should call `Fatalf` instead of `Errorf`.
Automatic merge from submit-queue (batch tested with PRs 50806, 48789, 49922, 49935, 50438)
iSCSI volume plugin: iSCSI initiatorname support
**What this PR does / why we need it**:
This PR adds iSCSI initiatorname parameter to ISCSIVolumeSource to enable automatic configuration of initiator name per volume. This would allow for more fine grained configuration, and remove the need to configure the initiator name on the host by administrator.
**Which issue this PR fixes**: fixes#47311
**Special notes for your reviewer**:
/cc @rootfs @jsafrane @msau42
**Release note**:
```
iSCSI volume plugin: Support customization of iSCSI initiator name per volume
```
This PR adds iSCSI initiatorname parameter to ISCSIVolumeSource
to enable automatic configuration of initiator name per volume.
This would allow for more fine grained configuration, and remove
the need to configure the initiator name on the host by
administrator.
fixes: #47311
Ben Swartzlander