It's unusual to put any kind of credentials out in open. The set
of credentials defined with `S3_ACCESS_KEY` and `S3_SECRET_KEY`
variables in the script are questionable in its nature. But these
are not real/valid credential values in any form for AWS rather
used for testing the S3 compatible API from Ceph RGW. Therefore
clarify the intention and replace with sample values from official
AWS documentation.
Signed-off-by: Anoop C S <anoopcs@cryptolab.net>
Workflow to automatically create a GitHub issue with a template
listing required API updates is in place for a while. Let's remove
the corresponding TODO from the file.
Signed-off-by: Anoop C S <anoopcs@cryptolab.net>
this commit renames ErrNoEntry of cephfs
to ErrNotExist and making it public allowing
other projects to access this.
Signed-off-by: Riya Singhal <rsinghal@redhat.com>
whether or not the allocation should be allowed to shrink, and a progress-tracking
callback. This contribution extends go-ceph to be able to call rbd_resize2.
Closes#933
Signed-off-by: Khayyam Saleem <khayyam.saleem@gmail.com>
Implement support for new /info AdminOps interface, which returns
a json record similar to:
{
"info": {
"storage_backends": [
{
"name":"rados",
"cluster_id":"204a1415-0b03-4926-bf76-6aebd192a52d"
}
]
}
}
Co-authored-by: Anoop C S <anoopcs@cryptolab.net>
Co-authored-by: Sébastien Han <seb@redhat.com>
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
Signed-off-by: Anoop C S <anoopcs@cryptolab.net>
Since Go 1.21 there is a runtime.Pinner API that allows to safely
pass structures with embedded Go pointers to C code. In earlier Go
version we know that the garbage collector is non-moving, so it is
safe to pass Go pointers to C as well. This change adds two
implementations of PtrGuard, one for pre 1.21 that is basically a
no-op, and one for 1.21+ that uses runtime.Pinner.
Signed-off-by: Sven Anderson <sven@redhat.com>
In my opinion the list of releases and the ceph versions supported was
getting a bit long. I am taking the initiative to split the list and
record the older releases in a separate doc. The cut point was a bit
arbitrary but I chose to remove all the caveats from the readme.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
By placing actions/checkout before actions/setup-go[1] we could avoid
the following warning message:
Warning: Restore cache failed: Dependencies file is not found in
/home/runner/work/go-ceph/go-ceph. Supported file pattern: go.sum
[1] https://github.com/actions/setup-go/issues/281
Signed-off-by: Anoop C S <anoopcs@cryptolab.net>
Enable stale bot to mark PRs stale and eventually close stale PRs. I had
tried to manually do this in the past with nice comments but I forget
to do it and trying to be nice was unrewarding. I find a bunch of old
PRs in the list a distraction and don't want to have to think about them
or try to follow up on them. I want to be generous to PR authors so PRs
can be inactive for 60 days and one has 3 weeks to unstale the PR.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
After adding the reef job, it was noticed that there was no quincy job.
Apparently we (probably me) forgot to add it at the time we added quincy
to our CI matrix, so we're adding it now.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
Currently the tests on the released reef image fails due to an RPM
transaction error when updating nfs-ganesha. Since go-ceph doesn't
need ganesha packages to be up-to-date to ensure that
lib{cephfs,rados,rbd}-devel are installed we _temporarily_ disable
ganesha repo when running yum update.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
As reef is being added and there's very little activity going into
pacific now, it should be safe to stop testing pre-pacific.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
Add CEPH_VERSION values for reef and pre-reef. This will allow building
and testing with ceph reef images.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
When a parent image has been removed, it will linger in the trash until
all siblings are gone. The image is not accessible through it's name
anymore, only through its ID.
The ImageSpec that is returned by Image.GetParent() now contains the
Trash boolean and the ImageID to identify if the image is in the trash,
and use OpenImageById() to access the removed parent image.
Related-to: ceph/ceph-csi#4013
Signed-off-by: Niels de Vos <ndevos@ibm.com>
Anoop C S
dependabot[bot]
Sven Anderson
Riya Singhal
Khayyam Saleem
John Mulligan
DanWritesCode
Dmitry Kvashnin
Niels de Vos
guoguangwu