From: Kit Westneat Subject: e2fsck can't handle bad extents w/ bad bitmaps Date: Tue, 10 Dec 2013 15:39:17 -0500 Message-ID: <52A77BF5.5060204@ddn.com> Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1"; format=flowed Content-Transfer-Encoding: 7bit To: "linux-ext4@vger.kernel.org" Return-path: Received: from legacy.ddn.com ([64.47.133.206]:25985 "EHLO legacy.ddn.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751197Ab3LJUjY (ORCPT ); Tue, 10 Dec 2013 15:39:24 -0500 Sender: linux-ext4-owner@vger.kernel.org List-ID: Hello, We ran into an issue recently where we had a filesystem that had both invalid bitmaps and corrupted extent nodes. The issue was that currently the fix for a corrupted extent requires that the bitmaps be read in: pass1.c +2250, scan_extent_node: pctx->blk = extent.e_pblk; pctx->blk2 = extent.e_lblk; pctx->num = extent.e_len; if (fix_problem(ctx, problem, pctx)) { >> e2fsck_read_bitmaps(ctx); pctx->errcode = ext2fs_extent_delete(ehandle, 0); Which then fails: e2fsck/util.c +272, e2fsck_read_bitmaps: if (ctx->invalid_bitmaps) { com_err(ctx->program_name, 0, _("e2fsck_read_bitmaps: illegal bitmap block(s) for %s"), We just ended up adding a patch to ignore the bad extents, but I just wanted to let people know that this problem exists. Thanks, Kit -- --- Kit Westneat L3 Lustre Support, DDN 703-659-3869