From mboxrd@z Thu Jan 1 00:00:00 1970 From: Erik Logtenberg Subject: Re: Kernel error during btrfs balance Date: Wed, 26 Jan 2011 10:43:09 +0100 Message-ID: <4D3FECAD.3080102@logtenberg.eu> References: <4D344EBB.6030903@logtenberg.eu> <4D3452C9.901@logtenberg.eu> <4D3948CB.8090801@logtenberg.eu> <4D3FE382.6070000@logtenberg.eu> <20110126092746.GJ29985@carfax.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 To: Hugo Mills , linux-btrfs@vger.kernel.org Return-path: In-Reply-To: <20110126092746.GJ29985@carfax.org.uk> List-ID: >> [104178.827624] entry offset 8891924480, bytes 4096, bitmap yes >> [104178.827626] block group has cluster?: no >> [104178.827628] 0 blocks of free space at or bigger than bytes is >> [104178.827631] block group 17213423616 has 5368709120 bytes, 5368709120 >> used 0 pinned 0 reserved >> [104178.827634] block group has cluster?: no >> >> And so on. >> >> Does this indicate an error of any sort, or is this expected behaviour? > > As far as I know, it means that you've run out of space, and not > every block group has been rewritten by the balance process. > > Hugo. > It is a 300GB volume with 79GB free. So hardly out of space. Moreover, I started the balance operation with the sole purpose of reclaiming some free space. The volume had like 40GB less free space when balance started, which was used by / reserved for Metadata. Kind regards, Erik.