1 |
On 1/4/07, maillists <lists@×××××.net> wrote: |
2 |
> On Thu, 2007-01-04 at 15:53 +0100, Jürgen Pierau wrote: |
3 |
> > maillists schrieb: |
4 |
> > > On Thu, 2007-01-04 at 12:32 +0100, Jürgen Pierau wrote: |
5 |
> > > |
6 |
> > >> maillists schrieb: |
7 |
> > >> |
8 |
> > >>> Hi, |
9 |
> > >>> |
10 |
> > >>> I just updated gcc and am running 4.1.1. but now my emerge does not work |
11 |
> > >>> at all. |
12 |
> > >>> |
13 |
> > >>> Here is an example of trying to update portage for example: |
14 |
> > >>> |
15 |
> > >>> thor rbragg # emerge -a portage |
16 |
> > >>> emergelog(): [Errno 30] Read-only file system: '/var/log/emerge.log' |
17 |
> > >>> emergelog(): [Errno 30] Read-only file system: '/var/log/emerge.log' |
18 |
> > >>> |
19 |
> > >>> |
20 |
> > >> It looks like your filesystem is mounted with the 'ro' option. Seeing |
21 |
> > >> the output of 'mount' would help. |
22 |
> > >> |
23 |
> > >> |
24 |
> > > Thanks, here is the "mount" command: As you can see I also have NFS |
25 |
> > > mount some local file shares from another host. My system is |
26 |
> > > on /dev/hda4. What is the "commit=0" thing? |
27 |
> > > |
28 |
> > > thor rbragg # mount |
29 |
> > > /dev/hda4 on / type ext3 (rw,commit=0) |
30 |
> > > |
31 |
> > Hi Rick, |
32 |
> > |
33 |
> > the mount line for your / looks ok to me (don't know about the commit, |
34 |
> > but afaik it shouldn't cause trouble). I'm somewhat at a loss. |
35 |
> > "Read-only file system" suggests that the file system driver (ext3 in |
36 |
> > this case) refuses to write to /dev/hda4. This should be reflected by a |
37 |
> > 'ro' (for read-only) in the mount line. Instead, there is 'rw' which is |
38 |
> > quite what it should be. It just doesn't fit the error message. |
39 |
> > |
40 |
> > I guess you have made sure that /var/log exists. Type 'dmesg' and check |
41 |
> > the output for messages concerning hda4 or ext3/ext2. Maybe the fs |
42 |
> > driver has decided to (take a) break for some reason (like fs errors, |
43 |
> > read errors or some such fancy occurrence). |
44 |
> > |
45 |
> > C'Ya, |
46 |
> > Jürgen |
47 |
> > |
48 |
> |
49 |
> Thanks, |
50 |
> |
51 |
> I decided to reboot (even though I unmerged pam-login) I was forced to |
52 |
> run fsck manually during boot and had alot of "orphaned inodes and tons |
53 |
> of other errors on the disk. Now it fails to start /etc/init.d/clock |
54 |
> and dies. Can't even boot! |
55 |
> |
56 |
> This whole thing started in the middle of updating glibc, and that never |
57 |
> finished. I was trying to update "emerge -auvDN system" and about 1/2 |
58 |
> hour into that, I got the first Read Only error... |
59 |
> |
60 |
> I'm hosed with the whole thing. Does anyone know when gentoo 2007.1 will |
61 |
> be released? |
62 |
> |
63 |
> thanks again for all the help! |
64 |
> rick |
65 |
|
66 |
If you haven't started reiinstalling yet, try this (I know I've done |
67 |
this on other distros, but to be honest, I haven't tried it with |
68 |
Gentoo): |
69 |
assuming you're using grub |
70 |
select your boot image and press 'e' for edit |
71 |
select the kernel line and, again, press 'e' for edit |
72 |
move to the end of the line and append: init=/bin/bash |
73 |
press return |
74 |
press 'b' for boot |
75 |
|
76 |
If all goes well, it should drop you directly into a bash command prompt. |
77 |
The root is probably mounted ro, so make it writable with mount -oremount,rw / |
78 |
>From there you should be able to use rc-update to remove clock from |
79 |
the boot run-level. |
80 |
Remount root as ro before exiting the shell (mount -o remount,ro /) |
81 |
|
82 |
dcm |
83 |
|
84 |
-- |
85 |
gentoo-laptop@g.o mailing list |