Re: [pve-devel] [PATCH pve-docs 3/3] fix #3967: add ZFS dRAID documentation

2022-06-03 Thread Thomas Lamprecht
Am 02/06/2022 um 14:47 schrieb Matthias Heiserer: >> >> +spares and data > I think we consistently start headers with a capital letter, so "spares" > should be "Spares" We actually use Title Case[0] for headings in all docs and also wiki articles, so it should be "Spares and Data" [0]: https://

Re: [pve-devel] [PATCH pve-storage 1/3] fix #3967: enable ZFS dRAID creation via API

2022-06-03 Thread Dominik Csapak
On 6/3/22 14:31, Dominik Csapak wrote: just saw an additional thing, comment inline On 6/2/22 13:22, Stefan Hrdlicka wrote: It is possible to set the number of spares and the size of data stripes via draidspares & dreaddata parameters. Signed-off-by: Stefan Hrdlicka ---   PVE/API2/Disks/ZFS.p

Re: [pve-devel] [PATCH pve-docs 3/3] fix #3967: add ZFS dRAID documentation

2022-06-03 Thread Dominik Csapak
comments inline On 6/2/22 13:22, Stefan Hrdlicka wrote: add some basic explanation how ZFS dRAID works including links to openZFS for more details add documentation for two dRAID parameters used in code Signed-off-by: Stefan Hrdlicka --- local-zfs.adoc | 40 +

Re: [pve-devel] [PATCH pve-storage 1/3] fix #3967: enable ZFS dRAID creation via API

2022-06-03 Thread Dominik Csapak
just saw an additional thing, comment inline On 6/2/22 13:22, Stefan Hrdlicka wrote: It is possible to set the number of spares and the size of data stripes via draidspares & dreaddata parameters. Signed-off-by: Stefan Hrdlicka --- PVE/API2/Disks/ZFS.pm | 40 +

Re: [pve-devel] [PATCH pve-manager 2/3] fix #3967: enable ZFS dRAID creation in WebGUI

2022-06-03 Thread Dominik Csapak
comments inline On 6/2/22 13:22, Stefan Hrdlicka wrote: add fields for additional settings required by ZFS dRAID Signed-off-by: Stefan Hrdlicka --- requires the changes in pve-storageto work www/manager6/node/ZFS.js | 47 1 file changed, 47 insertio

Re: [pve-devel] [PATCH pve-storage 1/3] fix #3967: enable ZFS dRAID creation via API

2022-06-03 Thread Dominik Csapak
some comments inline On 6/2/22 13:22, Stefan Hrdlicka wrote: It is possible to set the number of spares and the size of data stripes via draidspares & dreaddata parameters. Signed-off-by: Stefan Hrdlicka --- PVE/API2/Disks/ZFS.pm | 40 +++- 1 file changed

[pve-devel] applied series: [PATCH access-control 0/3] fix two propagation related bugs

2022-06-03 Thread Thomas Lamprecht
Am 03/06/2022 um 13:50 schrieb Fabian Grünbichler: > these patches fix two related bugs: > - the propagation flag used for priv dumping was set randomly if two > roles with a common priv exist on a path, one with and one without > propagation > - user/token priv intersection only took user priv

[pve-devel] [PATCH access-control 3/3] permissions: add some more comments

2022-06-03 Thread Fabian Grünbichler
for future reference/changes. Signed-off-by: Fabian Grünbichler --- src/PVE/RPCEnvironment.pm | 40 +-- 1 file changed, 38 insertions(+), 2 deletions(-) diff --git a/src/PVE/RPCEnvironment.pm b/src/PVE/RPCEnvironment.pm index a0c7555..08e39ff 100644 --- a/src

[pve-devel] [PATCH access-control 0/3] fix two propagation related bugs

2022-06-03 Thread Fabian Grünbichler
these patches fix two related bugs: - the propagation flag used for priv dumping was set randomly if two roles with a common priv exist on a path, one with and one without propagation - user/token priv intersection only took user privs into account that had propagation set the first can affe

[pve-devel] [PATCH access-control 1/3] permissions: properly merge propagation flag

2022-06-03 Thread Fabian Grünbichler
when multiple roles are defined on a path that share a privilege, this randomly took the propagation flag for the priv from the last role encountered. since perl hashes are iterated randomly, this means the propagation flag was sometimes set correctly, and sometimes not. note that this propagation

[pve-devel] [PATCH access-control 2/3] permissions: fix token/user priv intersection

2022-06-03 Thread Fabian Grünbichler
the token/user priv intersection could only honored user privs that had the propagation flag set, reducing the scope of the token more than intended. the pre-existing test case actually triggered the broken behaviour, but the expected value matched it so it was not noticed. Fixes: e8a0cee47bb4771

[pve-devel] [PATCH guest-common v2 2/2] ReplicationState: deterministically order replication jobs

2022-06-03 Thread Dominik Csapak
if we have multiple jobs for the same vmid with the same schedule, the last_sync, next_sync and vmid will always be the same, so the order depends on the order of the $jobs hash (which is random; thanks perl) to have a fixed order, take the jobid also into consideration Signed-off-by: Dominik Csa

[pve-devel] [PATCH guest-common v2 1/2] ReplicationState: purge state from non local vms

2022-06-03 Thread Dominik Csapak
when running replication, we don't want to keep replication states for non-local vms. Normally this would not be a problem, since on migration, we transfer the states anyway, but when the ha-manager steals a vm, it cannot do that. In that case, having an old state lying around is harmful, since the