Re: Can anyone tell me the meaning of this error and if I should

Thomas Schenk (tschenk@dejanews.com)
Wed, 18 Mar 1998 10:39:44 -0600 (CST)

On 18-Mar-98 Theodore Y. Ts'o wrote:
>
> Mar 17 14:06:35 dc7 kernel: EXT2-fs error (device 08:05): ext2_find_entry:
> directory #335873 contains a hole at offset 28672
>
> It's relatively harmless; e2fsck will fix it, and it doesn't harm your
> data. However, it's a condition which should never happen under normal
> operations using the current 2.0 and 2.1 kernels, so the fact that
> you're seeing it on your SMP systems is troubling.
>
> Would you be willing to run debugfs on that filesystem and use the
> command "stat <335873>" and tell me what debugfs prints? That would be
> most useful. Thanks!!

Here is the data:

dc7:~# debugfs /dev/sda5
debugfs: stat <335873>
Inode: 335873 Type: directory Mode: 0755 Flags: 0x0 Version: 1
User: 80 Group: 21 Size: 64512
File ACL: 0 Directory ACL: 0
Links: 0 Blockcount: 128
Fragment: Address: 0 Number: 0 Size: 0
ctime: 0x350fe8cc -- Wed Mar 18 09:31:24 1998
atime: 0x350fe8cc -- Wed Mar 18 09:31:24 1998
mtime: 0x350fe8cc -- Wed Mar 18 09:31:24 1998
dtime: 0x350fe8cc -- Wed Mar 18 09:31:24 1998
BLOCKS:
1343755 1346993 1347844 1349089 1349798 1351204 1352340 1352677
1352921 1353232 510 511 512 400283 1310748 808529971 858862391
858667315 926300216 825701176 400284 1376288 926495026 926364729
841888562 842215986 892613681 49 400285 1245212 825766193 876163126
858666295 825307191 3355700 400286 1376288 825308723 808925234
841889842 909588017 909521201 56 400287 1376288 875835441 943271990
825110833 859059761 892746036 57 400288 1310748 892810290 959460409
808726073 875902777 959788081 400289 1310748 909455666 892351028
858926645 859254839 926431283 400290 1310748 892418098 875902002
942551089 909129781 875902770 400291 1376288 892875058 808597305
858667056 842085936 960051001 50 400292 1310748 808858930 875902520
892221494 909129267 876099635 400293 1376288 926168882 875966776
875442744 842609202 825571381 52 400294 1376288 892875827 842477880
841887800 909326385 909456949 48 400295 1376288 909718322 892352305
858666032 825439542 825504561 54 400296 1245212 909717813 875836467
876162615 825571123 3487026 400297 1310748 892483633 960049462
942552375 892416569 825438257 400298 1376288 825636914 875639858
841889080 959525170 875640120 49 400299 1310748 926167602 859058741
942552630 875771697 875900979 400300 1310748 892416305 842545206
892481072 909193523 875705657 400301 1376288 842544947 909521968
825110580 959459638 909193272 48 400302 1310748 909392178 943011384
908998192 808597297 942879286 400303 1310748 825571385 926366003
942747192 892745271 875770419 400304 1245212 942945842 825766192
841889335 909588530 3158064 400305 1310748 909326130 825832753
909192759 875706161 825504565 400306 1310748 842019378 926431537
909258288 959788341 876163637 400307 1310748 925905203 842021173
858666808 943208761 960050995 400308 1310748 943142457 926167347
909258291 876165427 959722545 400309 1376288 909586996 842282544
825110833 808924977 926102580 57 400310 1376288 892548659 892940594
825112887 943142705 808859956 51 400311 1376288 808923698 892547636
858665010 959721529 909455924 48 400312 1310748 959657267 825570103
959329328 892877107 875705908 400313 1376288 959723058 875640118
825112885 926102329 942684208 51 400314 1376288 942813233 842543926
858665011 842151473 842281522 50 400315 1376288 859321905 942880050
858665267 808988727 875836212 55 400316 1179688 959590706 842084657
959655481 842479920 13107
TOTAL: 266

debugfs: quit

>
> The reason why that line has been #if 0'd out of the patch is because
> it's a relatively harmless case, and because in the future, it would be
> helpful to allow this case when removing pages from B-tree based
> directories.
>

Thanks for clearing that up. I also found another problem we have been
having that you might be able to shed some light on. On our news server,
we have been getting the following error messages:

Mar 17 10:17:36 nnrp2 kernel: EXT2-fs error (device 09:00): ext2_add_entry:
bad entry in directory #458756: rec_len is too small for name_len -
offset=5232, inode=6324492, rec_len=16, name_len=4101
Mar 17 10:17:36 nnrp2 kernel: EXT2-fs error (device 09:00): ext2_find_entry:
bad entry in directory #458756: rec_len is too small for name_len -
offset=5232, inode=6324492, rec_len=16, name_len=4101
Mar 17 10:17:36 nnrp2 kernel: EXT2-fs error (device 09:00): ext2_add_entry:
bad entry in directory #458756: rec_len is too small for name_len -
offset=5232, inode=6324492, rec_len=16, name_len=4101

Unfortunately, I can't send you debugfs output for this machine since the
news administrator rebuilt the filesystem before I could investigate.

I will keep watching for this error and if it occurs again, would be
willing to assist in whatever way I can in debugging this.

I truly appreciate you help with this.

Tom Schenk
Systems Administration Team Leader

---
+-----------------------+-------------------------+------------------------+
| Tom Schenk            | Use Linux!              | All opinions expressed |
| tschenk@dejanews.com  |    Friends don't let    | are mine and not those |
| tschenk@theoffice.net |     friends do Windows! | of my employer.        |
+-----------------------+-------------------------+------------------------+


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu