From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from firstgate.proxmox.com (firstgate.proxmox.com [212.224.123.68]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by lists.proxmox.com (Postfix) with ESMTPS id 484F0C5A4 for ; Tue, 29 Nov 2022 10:50:49 +0100 (CET) Received: from firstgate.proxmox.com (localhost [127.0.0.1]) by firstgate.proxmox.com (Proxmox) with ESMTP id 2C2ED5A18 for ; Tue, 29 Nov 2022 10:50:19 +0100 (CET) Received: from proxmox-new.maurer-it.com (proxmox-new.maurer-it.com [94.136.29.106]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by firstgate.proxmox.com (Proxmox) with ESMTPS for ; Tue, 29 Nov 2022 10:50:18 +0100 (CET) Received: from proxmox-new.maurer-it.com (localhost.localdomain [127.0.0.1]) by proxmox-new.maurer-it.com (Proxmox) with ESMTP id 247E4426D5 for ; Tue, 29 Nov 2022 10:50:18 +0100 (CET) Date: Tue, 29 Nov 2022 10:50:11 +0100 From: Fabian =?iso-8859-1?q?Gr=FCnbichler?= To: Fiona Ebner , pve-devel@lists.proxmox.com References: <20221107110035.93972-1-f.ebner@proxmox.com> <1668596522.lpeo4rqk2k.astroid@yuna.none> <39b12b91-05eb-8ad3-d7e0-6e67a3d1d103@proxmox.com> <174e5f00-af58-d440-6e87-a194526bc400@proxmox.com> In-Reply-To: <174e5f00-af58-d440-6e87-a194526bc400@proxmox.com> MIME-Version: 1.0 User-Agent: astroid/0.16.0 (https://github.com/astroidmail/astroid) Message-Id: <1669714828.ttb877ova7.astroid@yuna.none> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-SPAM-LEVEL: Spam detection results: 0 AWL 0.136 Adjusted score from AWL reputation of From: address BAYES_00 -1.9 Bayes spam probability is 0 to 1% KAM_DMARC_STATUS 0.01 Test Rule for DKIM or SPF Failure with Strict Alignment SPF_HELO_NONE 0.001 SPF: HELO does not publish an SPF Record SPF_PASS -0.001 SPF: sender matches SPF record Subject: Re: [pve-devel] [PATCH storage 1/2] zfs: only use cache when listing images locally X-BeenThere: pve-devel@lists.proxmox.com X-Mailman-Version: 2.1.29 Precedence: list List-Id: Proxmox VE development discussion List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 29 Nov 2022 09:50:49 -0000 On November 28, 2022 1:19 pm, Fiona Ebner wrote: > Am 16.11.22 um 14:30 schrieb Fiona Ebner: >> Am 16.11.22 um 12:18 schrieb Fabian Gr=C3=BCnbichler: >>> On November 7, 2022 12:00 pm, Fiona Ebner wrote: >>>> The plugin for remote ZFS storages currently also uses the same >>>> list_images() as the plugin for local ZFS storages. The issue with >>>> this is that there is only one cache which does not remember the >>>> target host where the information originated. >>>> >>>> Simply restrict the cache to be used for the local ZFS plugin only. An >>>> alternative solution would be to use a cache for each target host, but >>>> that seems a bit more involved and could still be added in the future. >>> >>> wouldn't it be sufficient to just do >>> >>> $cache->{zfs}->{$storeid} >>> >>> when filling/querying the cache, and combining that with *always* listi= ng only >>> the storage-relevant pool? >>=20 >> Yes, should work. I'll send a v2 with that. >>=20 >=20 > Well, a $storeid-based cache would be useless for both existing callers > using a cache parameter (pvesr's prepare_local_job and Storage.pm's > vdisk_list), because they iterate over the storages once for a given cach= e. the vdisk_list one is one-off actions only anyway, so there no cache or pos= sibly more overhead for multiple run_command executions is not that bad (compared= to the overhead of querying multiple pools while only being interested in one!= ). for replication, I guess in most cases this will be a single storage anyway= , unless we frequently pass in unrelated storages for scanning? > Should I get rid of the cache here entirely or do we go with this series > after all? judgment call I guess ;) > Also, I guess pvesr should switch to using Storage.pm's interface, > rather than talk to the plugin directly. yes. this should be a general rule (which is broken in some parts of our co= de atm) -> PVE::Storage is the interface to talk to storages, and that (and so= me other code in pve-storage) uses the plugins. might require some re-working though, but with other storage types up for replication in the near future putting some (more) parts behind an abstraction in pve-storage if needed se= ems sensible anyway ;)