linux-kernel - Re: FS-Cache: Duplicate cookie detected

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 12 Mar 2019 07:01:48 +0000
From:   David Howells <dhowells@...hat.com>
To:     Christian Kujau <lists@...dbynature.de>
Cc:     dhowells@...hat.com, linux-kernel@...r.kernel.org,
        Anna Schumaker <anna.schumaker@...app.com>,
        Steve Dickson <steved@...hat.com>
Subject: Re: FS-Cache: Duplicate cookie detected

Christian Kujau <lists@...dbynature.de> wrote:

> My /usr/local/src mount was mounted with vers=4.2 (default), while 
> nfstest_cache was mounting its test-mount with vers=4.1! Apart from the 
> different rsize/wsize values, the version number stood out. And indeed, 
> when I mount my regular NFS mount /usr/local/src with vers=4.1, the 
> "duplicate cookie" is no longer printed.

Yeah - NFS superblocks are differentiated by a whole host of parameters,
including protocol version number, and caches aren't shared between
superblocks because this introduces a tricky coherency problem.

The issue is that NFS superblocks to the same place do not currently manage
coherency (inode attributes, data) between themselves, except via the server.

However, if "fsc" isn't given on the mount commandline, the superblock
probably shouldn't get a server-level cookie if we can avoid it.

David