Re: [gentoo-user] openrc - network configuration fails - how to debug it
On 12/09/2012 05:50:35 PM, Neil Bothwick wrote: On Sun, 09 Dec 2012 15:18:49 +0100, Helmut Jarausch wrote: > config_wlan0="192.168.1.3 netmsk 255.255.255.0" ^ If this is a direct paste from your config, there's the problem. Thanks Neil for spotting this. It was the problem, indeed. Helmut.
[gentoo-user] Re: broadcom-sta and the 3.6.x kernel
On 2012-12-07, Mick wrote: > On Tuesday 04 Dec 2012 15:30:04 Dustin C. Hatch wrote: >> On 12/4/2012 06:11, Florian Philipp wrote: >> > Do you actually need broadcom-sta anymore? With the recent kernel >> > updates more chips work with the in-kernel driver (brcmsmac). But the >> > config option is well hidden (you need to enable BCMA to even see it). >> >> Yes, I initially tried the b43 driver, which worked, but consistently >> dropped about 5-15% of packets, making it mostly unusable. I also tried >> bcrmsmac and bcrmfmac, and neither of them supported my card (432b). >> Unfortunately, I can't get a different card, either, because I my >> notebook has a "whitelist" of supported devices in the BIOS, and it >> won't even boot with a mini-pci-e card installed that isn't in that >> list. Thanks, HP :( > > This sounds scary!!! Isn't there a way of disabling this feature in > the BIOS? With HP, you don't even get a BIOS setup. You get something that tells you the processor temperature and possibly lets you change the boot order. > Have you spoken to the HP police to ask what they can do to allow you to > manage the machine you bought from them? O_O I guess I could do that too. I find it a bit annoying that they don't even offer a BIOS setup and then decide to silently flip some of the settings with BIOS upgrades (like disabling AMD-V...) -- Nuno Silva (aka njsg) http://njsg.sdf-eu.org/
Re: [gentoo-user] Re: broadcom-sta and the 3.6.x kernel
On Tuesday 11 Dec 2012 09:04:59 Nuno J. Silva wrote: > On 2012-12-07, Mick wrote: > > On Tuesday 04 Dec 2012 15:30:04 Dustin C. Hatch wrote: > >> On 12/4/2012 06:11, Florian Philipp wrote: > >> > Do you actually need broadcom-sta anymore? With the recent kernel > >> > updates more chips work with the in-kernel driver (brcmsmac). But the > >> > config option is well hidden (you need to enable BCMA to even see it). > >> > >> Yes, I initially tried the b43 driver, which worked, but consistently > >> dropped about 5-15% of packets, making it mostly unusable. I also tried > >> bcrmsmac and bcrmfmac, and neither of them supported my card (432b). > >> Unfortunately, I can't get a different card, either, because I my > >> notebook has a "whitelist" of supported devices in the BIOS, and it > >> won't even boot with a mini-pci-e card installed that isn't in that > >> list. Thanks, HP :( > > > > This sounds scary!!! Isn't there a way of disabling this feature in > > the BIOS? > > With HP, you don't even get a BIOS setup. You get something that tells > you the processor temperature and possibly lets you change the boot > order. > > > Have you spoken to the HP police to ask what they can do to allow you to > > manage the machine you bought from them? O_O > > I guess I could do that too. I find it a bit annoying that they don't > even offer a BIOS setup and then decide to silently flip some of the > settings with BIOS upgrades (like disabling AMD-V...) Although I used to like Compaq (until they were bought out by HP) what you're describing is a place I really don't want to go! I'd be interested to hear what they come back with. Thanks for sharing. -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Localmount starts before LVM
On Tuesday 11 December 2012 01:14:39 PM IST, J. Roeleveld wrote:
>> Hi,
>>
>> I have a raid0 (kernel autodetect) array, over which I have put LVM
>> and then there are volumes on the LVM for /var, /tmp, swap and /home.
>>
>> The problem is, raid0 array gets recognized, but localmount fails to
>> mount because lvm doesn't seem to start before localmount (due to my
>> root being on SSD, I can't watch the output of openrc easily).
>>
>> For now I have added this to my rc.conf -
>> rc_localmount_before="lvm"
>
> In other words: "localmount" should run before "lvm"
>
>> rc_localmount_need="lvm"
>
> "localmount" requires "lvm"
>
>> rc_lvm_after="localmount"
>
> "lvm" should run after "localmount"
>
> Line 1 and 3 do the same. Line 2 is a contradiction.
>
>> This fixes the problem, but localmount still executes before lvm and
>> terminates with operational error. Then lvm starts up and localmount
>> runs again successfully.
>>
>> Any idea why this happens?
>
> Yes (See above)
>
>> The localmount script in init.d has proper depends:
>>
>> depend()
>> {
>> need fsck
>> use lvm modules mtab
>> after lvm modules
>> keyword -jail -openvz -prefix -vserver -lxc
>> }
>
> This should work.
>
> I actually have a similar setup and did not need to add the lines to rc.conf.
> All I did was do what I was told:
> Add "lvm" to the "boot" runlevel.
>
> Can you remove the lines from "rc.conf", ensure "lvm" is in the "boot"
> runlevel (And not in any other, like "default") and then let us know if
> you still get the error during reboot?
>
> If it all goes by too fast, can you press "I" during boot to get
> interactive and then let us know:
> 1) Which starts first, "lvm" or "localmount"
> 2) What error messages do you see for any of the services.
>
> Kind regards,
>
> Joost Roeleveld
>
>
Removing those lines didn't help, but I removed my stupidity there --
the contradicting dependency issue.
It still doesn't start up before localmount.
What I get when rc.conf is default without any manually inserted
depends/etc:
https://dl.dropbox.com/u/25780056/2012-12-11%2012.46.59.jpg &
https://dl.dropbox.com/u/25780056/2012-12-11%2012.48.13.jpg
My current rc.conf has this:
rc_localmount_need="lvm"
rc_localmount_after="lvm"
rc_fsck_after="lvm"
rc_fsck_need="lvm"
rc_lvm_before="localmount"
At least I have a usable system now and doesn't use my SSD for /var due
to the failed LVM mount.
But this results in one localmount failure -> lvm -> localmount success.
I'm on openrc 0.11.8.
--
Nilesh Govindarajan
http://nileshgr.com
Re: [gentoo-user] Localmount starts before LVM
> On Tuesday 11 December 2012 01:14:39 PM IST, J. Roeleveld wrote:
>>> Hi,
>>>
>>> I have a raid0 (kernel autodetect) array, over which I have put LVM
>>> and then there are volumes on the LVM for /var, /tmp, swap and /home.
>>>
>>> The problem is, raid0 array gets recognized, but localmount fails to
>>> mount because lvm doesn't seem to start before localmount (due to my
>>> root being on SSD, I can't watch the output of openrc easily).
>>>
>>> For now I have added this to my rc.conf -
>>> rc_localmount_before="lvm"
>>
>> In other words: "localmount" should run before "lvm"
>>
>>> rc_localmount_need="lvm"
>>
>> "localmount" requires "lvm"
>>
>>> rc_lvm_after="localmount"
>>
>> "lvm" should run after "localmount"
>>
>> Line 1 and 3 do the same. Line 2 is a contradiction.
>>
>>> This fixes the problem, but localmount still executes before lvm and
>>> terminates with operational error. Then lvm starts up and localmount
>>> runs again successfully.
>>>
>>> Any idea why this happens?
>>
>> Yes (See above)
>>
>>> The localmount script in init.d has proper depends:
>>>
>>> depend()
>>> {
>>> need fsck
>>> use lvm modules mtab
>>> after lvm modules
>>> keyword -jail -openvz -prefix -vserver -lxc
>>> }
>>
>> This should work.
>>
>> I actually have a similar setup and did not need to add the lines to
>> rc.conf.
>> All I did was do what I was told:
>> Add "lvm" to the "boot" runlevel.
>>
>> Can you remove the lines from "rc.conf", ensure "lvm" is in the "boot"
>> runlevel (And not in any other, like "default") and then let us know if
>> you still get the error during reboot?
>>
>> If it all goes by too fast, can you press "I" during boot to get
>> interactive and then let us know:
>> 1) Which starts first, "lvm" or "localmount"
>> 2) What error messages do you see for any of the services.
>>
>> Kind regards,
>>
>> Joost Roeleveld
>>
>>
>
> Removing those lines didn't help,
What is the end-result without the lines?
> but I removed my stupidity there --
> the contradicting dependency issue.
> It still doesn't start up before localmount.
>
> What I get when rc.conf is default without any manually inserted
> depends/etc:
> https://dl.dropbox.com/u/25780056/2012-12-11%2012.46.59.jpg &
> https://dl.dropbox.com/u/25780056/2012-12-11%2012.48.13.jpg
>
> My current rc.conf has this:
>
> rc_localmount_need="lvm"
> rc_localmount_after="lvm"
> rc_fsck_after="lvm"
> rc_fsck_need="lvm"
> rc_lvm_before="localmount"
>
> At least I have a usable system now and doesn't use my SSD for /var due
> to the failed LVM mount.
> But this results in one localmount failure -> lvm -> localmount success.
>
> I'm on openrc 0.11.8.
I use an older version still.
In rc.conf, I only set the "need" lines for init-scripts I created myself.
I never used the other lines.
Do you have /usr on / ? Or on a seperate partition?
Which metadata version did you use for the software raid setup?
Can you add "mdadm" to the boot-runlevel?
--
Joost
Re: [gentoo-user] Localmount starts before LVM
On Tuesday 11 December 2012 04:52 PM, J. Roeleveld wrote: > > What is the end-result without the lines? > > localmount fails at mounting /var /home and /tmp (while swap gets mounted which is *also* on LVM because lvm starts up before the swap gets activated). > > I use an older version still. > In rc.conf, I only set the "need" lines for init-scripts I created myself. > I never used the other lines. > I still have no idea why the sequence is messed up. I tried reverting to 0.10.5, but that didn't help either. > Do you have /usr on / ? Or on a seperate partition? > /usr is not a separate partition, it's on the same partition as root. > Which metadata version did you use for the software raid setup? > > Can you add "mdadm" to the boot-runlevel? > I'm using metadata version 1.2 for the raid0 array and the type is kernel based autodetect. Earlier I went by the raid guide on gentoo.org, but I configured it to use kernel based autodetect. mdadm anyway was reporting nothing detected (when added to boot runlevel) so it's not there in the boot runlevel. Moreover, since lvm starts up successfully, it doesn't seem to be an issue because of mdadm. It's just the sequence that's messed up. :S -- Nilesh Govindarajan http://nileshgr.com
Re: [gentoo-user] Localmount starts before LVM
> On Tuesday 11 December 2012 04:52 PM, J. Roeleveld wrote: >> Which metadata version did you use for the software raid setup? >> >> Can you add "mdadm" to the boot-runlevel? >> > > I'm using metadata version 1.2 for the raid0 array and the type is > kernel based autodetect. Ouch, auto-detect does not work with metadata 1.2. Please read the man-page section: === --auto-detect Request that the kernel starts any auto-detected arrays. This can only work if md is compiled into the kernel - not if it is a module. Arrays can be auto-detected by the kernel if all the components are in primary MS-DOS partitions with partition type FD, and all use v0.90 metadata. In-kernel autodetect is not recommended for new installations. Using mdadm to detect and assemble arrays - possibly in an initrd - is substantially more flexible and should be preferred. === > Earlier I went by the raid guide on gentoo.org, but I configured it to > use kernel based autodetect. > mdadm anyway was reporting nothing detected (when added to boot > runlevel) so it's not there in the boot runlevel. > > Moreover, since lvm starts up successfully, it doesn't seem to be an > issue because of mdadm. It's just the sequence that's messed up. :S Please rebuild the raid-device using v0.90 metadata and try again. Kind regards, Joost Roeleveld
[gentoo-user] ifconfig and ppp0 address
For years I have been using ifconfig and some simple shell magic to extract the ip address from the adsl ppp session when it changes. The latest update has changed the output format of ifconfig breaking things so if ifconfig cant be relied on, what's normally used to extract an interfaces IP address? I can easily rewrite the shell magic, but its worth asking if there is some command better suited. BillK
Re: [gentoo-user] ifconfig and ppp0 address
On Tue, 11 Dec 2012 19:33:15 +0800 Bill Kenworthy wrote: > For years I have been using ifconfig and some simple shell magic to > extract the ip address from the adsl ppp session when it changes. The > latest update has changed the output format of ifconfig breaking > things so if ifconfig cant be relied on, what's normally used to > extract an interfaces IP address? I can easily rewrite the shell > magic, but its worth asking if there is some command better suited. > > BillK > > > ip tends to be a little more consistent and a little more script-friendly than ifconfig: ip -o addr show -o prints output on one line, making your subsequent grep, sed and awking easier -- Alan McKinnon alan.mckin...@gmail.com
Re: [gentoo-user] Localmount starts before LVM
On Tue, 11 Dec 2012 12:48:13 +0100, J. Roeleveld wrote: > > I'm using metadata version 1.2 for the raid0 array and the type is > > kernel based autodetect. > > Ouch, auto-detect does not work with metadata 1.2. > Please read the man-page section: > > Please rebuild the raid-device using v0.90 metadata and try again. I don't understand why your using RAID at all. LVM on top of RAID0 makes no sense to me when you can simply make each device a PV and add it to the VG. That's more flexible and easier to repair. -- Neil Bothwick SITCOM: Single Income, Two Children, Oppressive Mortgage signature.asc Description: PGP signature
Re: [gentoo-user] ifconfig and ppp0 address
On 12/11/2012 12:33:15 PM, Bill Kenworthy wrote:
For years I have been using ifconfig and some simple shell magic to
extract the ip address from the adsl ppp session when it changes. The
latest update has changed the output format of ifconfig breaking
things
so if ifconfig cant be relied on, what's normally used to extract an
interfaces IP address? I can easily rewrite the shell magic, but its
worth asking if there is some command better suited.
If you're a Python fan there is a nice tool to replace 'awk' and
similar tools:
svn checkout http://pyp.googlecode.com/svn/trunk/ pyp
I've patched this to include a 'balanced' (parenthesis, brackets,..)
With the original version I use
ifconfig wlan0 | /usr/local/bin/pyp
"pp.after('wlan0',2)|p[1]|p.re(r'(?<=inet )\d+\.\d+\.\d+\.\d+')"
If someone is interested in my patch I'm happy to share it,
Helmut.
Re: [gentoo-user] Localmount starts before LVM
On Tue, 11 Dec 2012 12:08:12 + Neil Bothwick wrote: > On Tue, 11 Dec 2012 12:48:13 +0100, J. Roeleveld wrote: > > > > I'm using metadata version 1.2 for the raid0 array and the type is > > > kernel based autodetect. > > > > Ouch, auto-detect does not work with metadata 1.2. > > Please read the man-page section: > > > > Please rebuild the raid-device using v0.90 metadata and try again. > > I don't understand why your using RAID at all. LVM on top of RAID0 > makes no sense to me when you can simply make each device a PV and > add it to the VG. That's more flexible and easier to repair. > > Some folks like to do the striping in RAID, it's more controllable. 1st block on this disk, 2nd block on that disk, 3rd block on first disk again... Pooling LVM PVs into a VG is a huge gigantic basket of stuff where you don't really get to control very much - LVM sticks data wherever it wants to and you do little more than give some gentle hints (which I strongly suspect are mostly ignored) But yes, in the usual case RAID-0 on LVM doesn't make much sense for most folks. Personally, I prefer ZFS. This whole huge list of shit just goes away: disk partitions partition types disk labels worrying about if my block size is right worrying if my boundaries are correct PVs as different from VGs and LVs VGs as different from PVs and LVs LVs as different from PVs and VGs lvextend && growfs to make stuff bigger umount && shrinkfs && lvreduce && growfs && mount to make stuff smaller I can now take a much simpler view of things: "I have these disks, use 'em. When I've figured out the actual quotas and sizes I need, I'll let you know. Meanwhile just get on with it and store my stuff in some reasonable fashion, 'mkay? kthankxbye! I have real work to do." :-) -- Alan McKinnon alan.mckin...@gmail.com
Re: [gentoo-user] Localmount starts before LVM
On Tuesday 11 December 2012 05:18 PM, J. Roeleveld wrote: > > Ouch, auto-detect does not work with metadata 1.2. > Please read the man-page section: > === > --auto-detect > Request that the kernel starts any auto-detected arrays. This can only > work if md is compiled into the kernel - not if it is a module. Arrays > can be auto-detected by the kernel if all the components are in > primary MS-DOS partitions with partition type FD, and all use v0.90 > metadata. In-kernel autodetect is not recommended for new > installations. Using mdadm to detect and assemble arrays - possibly in > an initrd - is substantially more flexible and should be preferred. > === > > Please rebuild the raid-device using v0.90 metadata and try again. > I never had mdadm running in boot runlevel and I don't have a modular kernel. I have compiled everything into the kernel and hence no initrd either as I said earlier. Raid autodetection seems to work even _without_ mdadm running. -- [1.202481] md: Waiting for all devices to be available before autodetect [1.204268] md: If you don't use raid, use raid=noautodetect [1.206201] md: Autodetecting RAID arrays. [1.232482] md: invalid raid superblock magic on sdb1 [1.234306] md: sdb1 does not have a valid v0.90 superblock, not importing! [1.263187] md: invalid raid superblock magic on sdd1 [1.265034] md: sdd1 does not have a valid v0.90 superblock, not importing! [1.285106] md: invalid raid superblock magic on sdc1 [1.286960] md: sdc1 does not have a valid v0.90 superblock, not importing! [1.288787] md: Scanned 3 and added 0 devices. [1.290590] md: autorun ... [1.292380] md: ... autorun DONE. [1.340838] UDF-fs: warning (device sda1): udf_fill_super: No partition found (1) [1.350473] XFS (sda1): Mounting Filesystem [1.454096] usb usb5: suspend_rh (auto-stop) [1.454130] usb usb4: suspend_rh (auto-stop) [1.455673] usb usb2: suspend_rh (auto-stop) [1.455698] usb usb3: suspend_rh (auto-stop) [1.573933] XFS (sda1): Ending clean mount [1.575762] VFS: Mounted root (xfs filesystem) readonly on device 8:1. [1.578193] Freeing unused kernel memory: 456k freed [1.580154] BFS CPU scheduler v0.425 by Con Kolivas. [2.503599] systemd-udevd[974]: starting version 196 [2.704048] hub 2-0:1.0: hub_suspend [2.704063] usb usb2: bus auto-suspend, wakeup 1 [2.704068] usb usb2: suspend_rh [2.704091] hub 3-0:1.0: hub_suspend [2.704098] usb usb3: bus auto-suspend, wakeup 1 [2.704102] usb usb3: suspend_rh [2.708031] hub 4-0:1.0: hub_suspend [2.708041] usb usb4: bus auto-suspend, wakeup 1 [2.708046] usb usb4: suspend_rh [2.712023] hub 5-0:1.0: hub_suspend [2.712030] usb usb5: bus auto-suspend, wakeup 1 [2.712034] usb usb5: suspend_rh [2.794061] hub 1-6:1.0: hub_suspend [2.794072] usb 1-6: unlink qh256-0001/8800bb832980 start 1 [1/0 us] [2.797202] usb 1-6: usb auto-suspend, wakeup 1 [2.973953] md: bind [3.020879] md: bind [3.086724] md: bind [3.087690] bio: create slab at 1 [3.087705] md/raid0:md0: md_size is 2266111488 sectors. [3.087708] md: RAID0 configuration for md0 - 3 zones [0.524821] ACPI: Invalid Power Resource to register! [3.087711] md: zone0=[ [3.087714] sdb1/sdc1/sdd1] [3.087721] zone-offset= 0KB, device-offset= 0KB, size= 468863328KB [3.087723] md: zone1=[sdb1/sdc1] [3.087730] zone-offset= 468863328KB, device-offset= 156287776KB, size= 664191360KB [3.087732] md: zone2=[sdb1] [3.087737] zone-offset=1133054688KB, device-offset= 488383456KB, size= 1056KB [3.087752] md0: detected capacity change from 0 to 1160249081856 [3.098375] md0: unknown partition table - Where did md0 come from if it was not setup by mdadm? -- Nilesh Govindarajan http://nileshgr.com
Re: [gentoo-user] Localmount starts before LVM
On Tuesday 11 December 2012 05:57 PM, Alan McKinnon wrote: > On Tue, 11 Dec 2012 12:08:12 + > Neil Bothwick wrote: > >> On Tue, 11 Dec 2012 12:48:13 +0100, J. Roeleveld wrote: >> I'm using metadata version 1.2 for the raid0 array and the type is kernel based autodetect. >>> >>> Ouch, auto-detect does not work with metadata 1.2. >>> Please read the man-page section: >>> >>> Please rebuild the raid-device using v0.90 metadata and try again. >> >> I don't understand why your using RAID at all. LVM on top of RAID0 >> makes no sense to me when you can simply make each device a PV and >> add it to the VG. That's more flexible and easier to repair. >> >> > > Some folks like to do the striping in RAID, it's more controllable. 1st > block on this disk, 2nd block on that disk, 3rd block on first disk > again... > > Pooling LVM PVs into a VG is a huge gigantic basket of stuff where you > don't really get to control very much - LVM sticks data wherever it > wants to and you do little more than give some gentle hints (which > I strongly suspect are mostly ignored) > > But yes, in the usual case RAID-0 on LVM doesn't make much sense for > most folks. > > Personally, I prefer ZFS. This whole huge list of shit just goes away: > > disk partitions > partition types > disk labels > worrying about if my block size is right > worrying if my boundaries are correct > PVs as different from VGs and LVs > VGs as different from PVs and LVs > LVs as different from PVs and VGs > lvextend && growfs to make stuff bigger > umount && shrinkfs && lvreduce && growfs && mount to make stuff smaller > > I can now take a much simpler view of things: > > "I have these disks, use 'em. When I've figured out the actual quotas > and sizes I need, I'll let you know. Meanwhile just get on with it and > store my stuff in some reasonable fashion, 'mkay? kthankxbye! I have > real work to do." > > :-) > > Exactly the reason why I wanted RAID0 and LVM in combination: more IOPS. ZFS looks very interesting, how stable is it? -- Nilesh Govindarajan http://nileshgr.com
Re: [gentoo-user] Localmount starts before LVM
> On Tuesday 11 December 2012 05:18 PM, J. Roeleveld wrote: >> >> Ouch, auto-detect does not work with metadata 1.2. >> Please read the man-page section: >> === >> --auto-detect >> Request that the kernel starts any auto-detected arrays. This can >> only >> work if md is compiled into the kernel - not if it is a module. Arrays >> can be auto-detected by the kernel if all the components are in >> primary MS-DOS partitions with partition type FD, and all use v0.90 >> metadata. In-kernel autodetect is not recommended for new >> installations. Using mdadm to detect and assemble arrays - possibly in >> an initrd - is substantially more flexible and should be preferred. >> === >> >> Please rebuild the raid-device using v0.90 metadata and try again. >> > > I never had mdadm running in boot runlevel and I don't have a modular > kernel. I have compiled everything into the kernel and hence no initrd > either as I said earlier. > > Raid autodetection seems to work even _without_ mdadm running. > > -- > > [1.202481] md: Waiting for all devices to be available before > autodetect > [1.204268] md: If you don't use raid, use raid=noautodetect > [1.206201] md: Autodetecting RAID arrays. > [1.232482] md: invalid raid superblock magic on sdb1 > [1.234306] md: sdb1 does not have a valid v0.90 superblock, not > importing! > [1.263187] md: invalid raid superblock magic on sdd1 > [1.265034] md: sdd1 does not have a valid v0.90 superblock, not > importing! > [1.285106] md: invalid raid superblock magic on sdc1 > [1.286960] md: sdc1 does not have a valid v0.90 superblock, not > importing! > [1.288787] md: Scanned 3 and added 0 devices. This clearly indicates that the "autostart" is not working > [3.087705] md/raid0:md0: md_size is 2266111488 sectors. > [3.087708] md: RAID0 configuration for md0 - 3 zones > [0.524821] ACPI: Invalid Power Resource to register! > [3.087711] md: zone0=[ > [3.087714] sdb1/sdc1/sdd1] > [3.087721] zone-offset= 0KB, device-offset= > 0KB, size= 468863328KB > [3.087723] md: zone1=[sdb1/sdc1] > [3.087730] zone-offset= 468863328KB, device-offset= > 156287776KB, size= 664191360KB > [3.087732] md: zone2=[sdb1] > [3.087737] zone-offset=1133054688KB, device-offset= > 488383456KB, size= 1056KB > > [3.087752] md0: detected capacity change from 0 to 1160249081856 > [3.098375] md0: unknown partition table Something found and started md0 after the autoraid-detect clearly failed. > Where did md0 come from if it was not setup by mdadm? What does "rc-status" show right after boot? -- Joost
Re: [gentoo-user] Localmount starts before LVM
On Dec 11, 2012 7:57 AM, "Nilesh Govindrajan" wrote: > > On Tuesday 11 December 2012 05:18 PM, J. Roeleveld wrote: > > > > Ouch, auto-detect does not work with metadata 1.2. > > Please read the man-page section: > > === > > --auto-detect > > Request that the kernel starts any auto-detected arrays. This can only > > work if md is compiled into the kernel - not if it is a module. Arrays > > can be auto-detected by the kernel if all the components are in > > primary MS-DOS partitions with partition type FD, and all use v0.90 > > metadata. In-kernel autodetect is not recommended for new > > installations. Using mdadm to detect and assemble arrays - possibly in > > an initrd - is substantially more flexible and should be preferred. > > === > > > > Please rebuild the raid-device using v0.90 metadata and try again. > > > > I never had mdadm running in boot runlevel and I don't have a modular > kernel. I have compiled everything into the kernel and hence no initrd > either as I said earlier. > > Raid autodetection seems to work even _without_ mdadm running. > > -- > > [1.202481] md: Waiting for all devices to be available before autodetect > [1.204268] md: If you don't use raid, use raid=noautodetect > [1.206201] md: Autodetecting RAID arrays. > [1.232482] md: invalid raid superblock magic on sdb1 > [1.234306] md: sdb1 does not have a valid v0.90 superblock, not > importing! > [1.263187] md: invalid raid superblock magic on sdd1 > [1.265034] md: sdd1 does not have a valid v0.90 superblock, not > importing! > [1.285106] md: invalid raid superblock magic on sdc1 > [1.286960] md: sdc1 does not have a valid v0.90 superblock, not > importing! > [1.288787] md: Scanned 3 and added 0 devices. > [1.290590] md: autorun ... > [1.292380] md: ... autorun DONE. > [1.340838] UDF-fs: warning (device sda1): udf_fill_super: No > partition found (1) > [1.350473] XFS (sda1): Mounting Filesystem > [1.454096] usb usb5: suspend_rh (auto-stop) > [1.454130] usb usb4: suspend_rh (auto-stop) > [1.455673] usb usb2: suspend_rh (auto-stop) > [1.455698] usb usb3: suspend_rh (auto-stop) > [1.573933] XFS (sda1): Ending clean mount > [1.575762] VFS: Mounted root (xfs filesystem) readonly on device 8:1. > [1.578193] Freeing unused kernel memory: 456k freed > [1.580154] BFS CPU scheduler v0.425 by Con Kolivas. > [2.503599] systemd-udevd[974]: starting version 196 > [2.704048] hub 2-0:1.0: hub_suspend > [2.704063] usb usb2: bus auto-suspend, wakeup 1 > [2.704068] usb usb2: suspend_rh > [2.704091] hub 3-0:1.0: hub_suspend > [2.704098] usb usb3: bus auto-suspend, wakeup 1 > [2.704102] usb usb3: suspend_rh > [2.708031] hub 4-0:1.0: hub_suspend > [2.708041] usb usb4: bus auto-suspend, wakeup 1 > [2.708046] usb usb4: suspend_rh > [2.712023] hub 5-0:1.0: hub_suspend > [2.712030] usb usb5: bus auto-suspend, wakeup 1 > [2.712034] usb usb5: suspend_rh > [2.794061] hub 1-6:1.0: hub_suspend > [2.794072] usb 1-6: unlink qh256-0001/8800bb832980 start 1 [1/0 us] > [2.797202] usb 1-6: usb auto-suspend, wakeup 1 > [2.973953] md: bind > [3.020879] md: bind > [3.086724] md: bind > [3.087690] bio: create slab at 1 > [3.087705] md/raid0:md0: md_size is 2266111488 sectors. > [3.087708] md: RAID0 configuration for md0 - 3 zones > [0.524821] ACPI: Invalid Power Resource to register! > [3.087711] md: zone0=[ > [3.087714] sdb1/sdc1/sdd1] > [3.087721] zone-offset= 0KB, device-offset= > 0KB, size= 468863328KB > [3.087723] md: zone1=[sdb1/sdc1] > [3.087730] zone-offset= 468863328KB, device-offset= > 156287776KB, size= 664191360KB > [3.087732] md: zone2=[sdb1] > [3.087737] zone-offset=1133054688KB, device-offset= > 488383456KB, size= 1056KB > > [3.087752] md0: detected capacity change from 0 to 1160249081856 > [3.098375] md0: unknown partition table > > - > > Where did md0 come from if it was not setup by mdadm? Metadata format 0.9 supports auto-detection by the kernel. > > -- > Nilesh Govindarajan > http://nileshgr.com >
Re: [gentoo-user] System maintenance procedure?
On Sun, Dec 09, 2012 at 04:48:24PM +, Neil Bothwick wrote: > > > workstation ~ # emerge -a @preserved-rebuild > > emerge: 'preserved-rebuild' is an empty set > > emerge: no targets left after set expansion > > So you have nothing that needs rebuilding. Portage will warn you when the > set it non-empty, telling you to run emerge @preserved-rebuild. There is > no need to run it at any other time. After using Gentoo for close to two years, the only time/place I've ever even seen @preserved-rebuild is in this thread. Yet you say, "Portage will warn you when the set is [it] non-empty, telling you to run emerge @preserved-rebuild." How will portage do this? An alias 'ud' alias ud='eix-sync && emerge -aDjNuv @world && dispatch-conf && emerge -a --depclean && revdep-rebuild -i && clear && exit' is run every morning with my first cup of coffee. If something were changed or left off that alias do you suppose this mysterious @preserved-rebuild would be run? Thanks, Bruce -- Happy Penguin Computers >') 126 Fenco Drive ( \ Tupelo, MS 38801 ^^ supp...@happypenguincomputers.com 662-269-2706 662-205-6424 http://happypenguincomputers.com/ Don't top-post: http://en.wikipedia.org/wiki/Top_post#Top-posting
Re: [gentoo-user] Localmount starts before LVM
On Tue, 11 Dec 2012 18:28:37 +0530 Nilesh Govindrajan wrote: > > "I have these disks, use 'em. When I've figured out the actual > > quotas and sizes I need, I'll let you know. Meanwhile just get on > > with it and store my stuff in some reasonable fashion, 'mkay? > > kthankxbye! I have real work to do." > > > > :-) > > > > > > Exactly the reason why I wanted RAID0 and LVM in combination: more > IOPS. ZFS looks very interesting, how stable is it? On Linux, not at all (it doesn't exist there except using fuse) On FreeBSD, rock solid. On Solaris, rock solid. It almost seems to be everything btrfs is not... -- Alan McKinnon alan.mckin...@gmail.com
Re: [gentoo-user] System maintenance procedure?
On Tue, 11 Dec 2012 07:36:10 -0600, Bruce Hill wrote: > > So you have nothing that needs rebuilding. Portage will warn you when > > the set is non-empty, telling you to run emerge @preserved-rebuild. > > There is no need to run it at any other time. > > After using Gentoo for close to two years, the only time/place I've > ever even seen @preserved-rebuild is in this thread. Yet you say, > "Portage will warn you when the set is non-empty, telling you to > run emerge @preserved-rebuild." > > How will portage do this? An alias 'ud' A message printed to the terminal at the end of the emerge run... > alias ud='eix-sync && emerge -aDjNuv @world && dispatch-conf && emerge > -a --depclean && revdep-rebuild -i && clear && exit' which is hidden by the output from the subsequent commands. > is run every morning with my first cup of coffee. If something were > changed or left off that alias do you suppose this mysterious > @preserved-rebuild would be run? It is run when you choose to run it, it never happens automatically (unless there is a flag I don't know about). You could drop "emerge -a @preserved-rebuild" into the alias, between the emerge world and dispatch-conf. There's no harm in running it when it is not needed. % sudo emerge -a @preserved-rebuild emerge: 'preserved-rebuild' is an empty set emerge: no targets left after set expansion -- Neil Bothwick Why is bra singular and pants plural? signature.asc Description: PGP signature
[gentoo-user] Re: Localmount starts before LVM [resolved]
Issue resolved guys. Thanks to J. Roeleveld for pointing out that auto detection is failing. I built an initramfs using genkernel with raid and lvm support and added options domdadm & lvmraid. It works flawlessly now. Thanks! :-) -- Nilesh Govindarajan http://nileshgr.com
Re: [gentoo-user] ifconfig and ppp0 address
On Tue, 11 Dec 2012 19:33:15 +0800, Bill Kenworthy wrote about "[gentoo-user] ifconfig and ppp0 address": >For years I have been using ifconfig and some simple shell magic to >extract the ip address from the adsl ppp session when it changes. The >latest update has changed the output format of ifconfig breaking things >so if ifconfig cant be relied on, what's normally used to extract an >interfaces IP address? I can easily rewrite the shell magic, but its >worth asking if there is some command better suited. I wrote a command in C++ that interrogates the TCP/IP stack directly to obtain the IP address of any given interface. I wrote it some years ago, so it's only IPv4 at the moment; it will take a day or two to add IPv6 support. I can send you the source code if you want. Likewise to any other interested reader -- or I can post it here as a mail attachment, since it is quite tiny. -- Regards, Dave [RLU #314465] == dwn...@ntlworld.com (David W Noon) == signature.asc Description: PGP signature
Re: [gentoo-user] Localmount starts before LVM
Alan McKinnon wrote: On Tue, 11 Dec 2012 18:28:37 +0530 Nilesh Govindrajan wrote: Exactly the reason why I wanted RAID0 and LVM in combination: more IOPS. ZFS looks very interesting, how stable is it? On Linux, not at all (it doesn't exist there except using fuse) On FreeBSD, rock solid. On Solaris, rock solid. It almost seems to be everything btrfs is not... The details why this is the case are something I can never remember straight in my head, but I recall that it's due to licensing that ZFS cannot be included in the Linux kernel directly. I think it might be because the ZFS license doesn't have the Copyleft clause that the GPL requires? It's sad, because ZFS is really pretty great. I think btrfs will be pretty great too once it is stabilized, so I look forward to that. Also, I had seen some kernel patches that you can apply yourself to get ZFS in Linux without FUSE a year or two back. I never tried them, and can't attest to how stable or unstable they might be, but you could look into that as well. -- R
Re: [gentoo-user] System maintenance procedure?
On 12/11/2012 08:36 AM, Bruce Hill wrote: > On Sun, Dec 09, 2012 at 04:48:24PM +, Neil Bothwick wrote: >> >>> workstation ~ # emerge -a @preserved-rebuild >>> emerge: 'preserved-rebuild' is an empty set >>> emerge: no targets left after set expansion >> >> So you have nothing that needs rebuilding. Portage will warn you when the >> set it non-empty, telling you to run emerge @preserved-rebuild. There is >> no need to run it at any other time. > > After using Gentoo for close to two years, the only time/place I've ever even > seen @preserved-rebuild is in this thread. Yet you say, "Portage will warn you > when the set is [it] non-empty, telling you to run emerge @preserved-rebuild." > It's probably safe to pretend it doesn't exist now. We have a better solution in EAPI5 -- packages can force their dependents to rebuild after an upgrade. It will take a while to transition the whole tree, but if you remember to complain loudly whenever a libfoo upgrade breaks something, it will happen. In the meantime, run revdep-rebuild every once in a while.
[gentoo-user] Intel Atom: architecture, distcc, crossdev and compile flags
Hello list Long time no read... :) It follows a verbose preamble. For the actual questions see dashed line below. TL;DR summary: it’s all about ricer-performance questions on a netbook. I have the luck of having obtained a used netbook for free (Atom N450, single- core with HT, 1 GB memory, 5400 rev HDD). During the last week I’ve been experimenting with 32 and 64 bits on it and am still quite undecided which to keep. My reasons: - They are not as far apart in CPU performance as is the Core2. I posted a 32/64 comparison for Core2 a few months ago, which showed that Lilypond speedup on 64 bit was 50%. On the Atom, it actually took 5% longer. (Sadly, Blender doesn’t build on 32 bit right now). - Startup times for hogs like Firefox and KDE are quite equal between the two (that could be attributed in parts to the fact that the 64 bit partition sits on the disk’s first sectors, while 32 bit sits at the other end, I don’t know which end is faster). - pro 64: it is very easy to use distcc, as opposed to 32 bits (see below). - con 64: it uses about 50% more memory, 32 bit builds are a little faster. The RAM argument is the most convincing one right now, since more free RAM means more cache, which means a faster system in the long run. Currently, KDE after logon needs 150 MB on 32 bit, and 250 MB on 64 bit (without akonadi for now). But awesome WM rocks on a netbook anyway. --[ Questions begin ] So I’m interested in you opinion and own experience about the following arising questions: * From my observations, the benefit of 64 bit over 32 is much smaller for an Atom than it is for my Core2. Am I right to assume thus that the Atom architecture doesn’t have much to offer to 64 bit (such as extra registers)? I’m not talking about memory here, since it’s limited to 2 GB in any case. * The problem of distcc between different architectures: The netbook already had an older 32 bit Gentoo installed. And since I have a multilib host (march=core2), I though I could upgrade with distcc (using march=atom on the netbook). But at some point more and more stuff stopped working, eventually I got “invalid instruction” errors during emerge, hence I figured that was a dead end. So is it possible to mix architectures in this way at all with distcc? I also have crossdev for i686 installed, which even shares files with the system’s normal multilib gcc. I find that odd. I sped up the installation process for 32 bit by using a chroot on the big machine, which worked nicely. But it’s not a long-term solution, b/c it uses up too much disk space on the host. * I’m interested in the question of -O2 vs. -Os. Some sources say -Os is bad, b/c it breaks debugging and is mainly untested. I won’t do heavy developing on it anyway, and Atoms do have a puny cache. So I wonder whether -Os would improve execution time and RAM usage noticably. Diskspace itself is not an issue. * I’m also interested in comparing bin packages over self-compiled ones. E.g. I did compile icedtea, even if it’s just for TV browser. :) Can you name a Java benchmark to measure CPU performance? * The last thing I’m going to set up is filesystem encryption, at least for ~. I already know/think that AES would be the best choice due to limited CPU power, but what else is there to heed besides key size? * What other small benchmarks for CPU and memory can you recommend? So far I tested with nbench and sysbench. The results are so-and-so. Some computation stuff is much slower on 64 bit, some a bit faster. The applicability to every-day use is of course a wibbly-wobbly argument. I also tested the runtime of some application (packing and unpacking of archives, throughput with dd, mencoder). If there is interest, I can post the result of 21 runs on each platform, measured with GNU time. --[ Questions end ]-- PS.: I’m aware that benchmarks are always a bit subjective and none is perfect. I also realise that most of the questions quite belong into the ricer corner. But Netbooks are ricer devices, b/c they need to perform at their limits all the time. :-D If you read until this point, thank you very much for your time. You get a unicorn as a reward: `^nn~ -- Gruß | Greetings | Qapla’ Please do not share anything from, with or about me with any Facebook service. “Time is money” said the waiter and put the date on the bill. signature.asc Description: Digital signature
[gentoo-user] Re: OT: first water cooled system
Volker Armin Hemmann googlemail.com> writes: > Then use burnp6. Or burnmmx. Same packacke. No winecrap. OK thanks for the input. Thank for all the help. James
Re: [gentoo-user] ifconfig and ppp0 address
> I can send you the source code if you want. Likewise to any other > interested reader Send to me please, Thanks -- ___ 'Write programs that do one thing and do it well. Write programs to work together. Write programs to handle text streams, because that is a universal interface' (Doug McIlroy) ___
Re: [gentoo-user] Re: broadcom-sta and the 3.6.x kernel
On 12/11/2012 03:04, Nuno J. Silva wrote: On 2012-12-07, Mick wrote: This sounds scary!!! Isn't there a way of disabling this feature in the BIOS? With HP, you don't even get a BIOS setup. You get something that tells you the processor temperature and possibly lets you change the boot order. That's exactly the case with this notebook. No settings whatsoever beyond date/time and boot order. Have you spoken to the HP police to ask what they can do to allow you to manage the machine you bought from them? O_O I guess I could do that too. I find it a bit annoying that they don't even offer a BIOS setup and then decide to silently flip some of the settings with BIOS upgrades (like disabling AMD-V...) This machine is long since out of warranty; they won't even speak to me anymore. I was able to use a hex editor to modify the white list at one time, but it didn't matter because the "hardware" radio switch doesn't work with different cards, either, leaving the radio in a permanently off state. Someday, I'll get a more professional notebook, but until then, I'm stuck with the patched broadcom-sta. It does seem to be working fine for now, so I'm not terribly worried. As much as I would prefer to have an open source driver, I also need my machine to work. -- ♫Dustin
Re: [gentoo-user] Installing specific layman packages & eix-test-obsolete
> > If my package.mask is empty, eix-test-obsolete runs fine. If I have > > this in package.mask: > > > > */*::init6 > > > > eix-test-obsolete find over 27,000 packages under this heading: > > > > Redundant in /etc/portage/package.mask: > > ... considered as REDUNDANT_IF_MASK_NO_CHANGE > > > > Adding the following to package.nowarn doesn't seem to change anything: > > > > */*::init6 mask_no_change > > > > The first of the 27,000 packages is app-accessibility/SphinxTrain which > > is in the portage tree and not in an overlay. > > > > Try a hammer: > > */*::init6 in_keywords no_change double mixed weaker double_line > in_mask mask_no_change double_masked in_unmask unmask_no_change > double_unmasked in_use double_use in_env double_env in_license > double_license in_cflags double_cflags nonexistent masked > other_overlay > > With nothing in package.nowarn, I get 700-some REDUNDANT_IF_NO_CHANGE > warnings. If I add just, > > */*::gentoo-haskell no_change > > I get output that looks almost exactly like the REDUNDANT_IF_NO_CHANGE > warnings, but if I look closely, it's for in-keywords-but-not-installed. > In other words, that single nowarn entry may be working, but you might > be getting some other warning now. > > If adding them all to package.nowarn makes the warnings go away, then > you just have to figure out which subset you need. I double-checked and if I have */*::init6 in package.mask, I get 27,000+ REDUNDANT_IF_MASK_NO_CHANGE warnings from eix-test-obsolete, with and without the following in package.nowarn: */*::init6 in_keywords no_change double mixed weaker double_line in_mask mask_no_change double_masked in_unmask unmask_no_change double_unmasked in_use double_use in_env double_env in_license double_license in_cflags double_cflags nonexistent masked other_overlay No other overlays installed. - Grant
Re: [gentoo-user] Intel Atom: architecture, distcc, crossdev and compile flags
On 11 December 2012 12:36, Frank Steinmetzger wrote: > Hello list . . . > So I’m interested in you opinion and own experience about the following > arising questions: > > * From my observations, the benefit of 64 bit over 32 is much smaller for an > Atom than it is for my Core2. Am I right to assume thus that the Atom > architecture doesn’t have much to offer to 64 bit (such as extra registers)? > I’m not talking about memory here, since it’s limited to 2 GB in any case. > > * The problem of distcc between different architectures: > The netbook already had an older 32 bit Gentoo installed. And since I have > a multilib host (march=core2), I though I could upgrade with distcc (using > march=atom on the netbook). But at some point more and more stuff stopped > working, eventually I got “invalid instruction” errors during emerge, hence > I figured that was a dead end. > > So is it possible to mix architectures in this way at all with distcc? > I also have crossdev for i686 installed, which even shares files with the > system’s normal multilib gcc. I find that odd. > I sped up the installation process for 32 bit by using a chroot on the big > machine, which worked nicely. But it’s not a long-term solution, b/c it > uses up too much disk space on the host. > > * I’m interested in the question of -O2 vs. -Os. > Some sources say -Os is bad, b/c it breaks debugging and is mainly untested. > I won’t do heavy developing on it anyway, and Atoms do have a puny cache. > So I wonder whether -Os would improve execution time and RAM usage > noticably. Diskspace itself is not an issue. > > * I’m also interested in comparing bin packages over self-compiled ones. > E.g. I did compile icedtea, even if it’s just for TV browser. :) > Can you name a Java benchmark to measure CPU performance? > > * The last thing I’m going to set up is filesystem encryption, at least for ~. > I already know/think that AES would be the best choice due to limited CPU > power, but what else is there to heed besides key size? > > * What other small benchmarks for CPU and memory can you recommend? So far I > tested with nbench and sysbench. The results are so-and-so. Some computation > stuff is much slower on 64 bit, some a bit faster. The applicability to > every-day use is of course a wibbly-wobbly argument. > I also tested the runtime of some application (packing and unpacking of > archives, throughput with dd, mencoder). If there is interest, I can post > the result of 21 runs on each platform, measured with GNU time. > > --[ Questions end ]-- > > > PS.: I’m aware that benchmarks are always a bit subjective and none is > perfect. I also realise that most of the questions quite belong into the > ricer corner. But Netbooks are ricer devices, b/c they need to perform at > their limits all the time. :-D > I have an old N280 atom netbook, so the 64v32 is moot for me, but with a hardware limit of 2G, I'd probably run 32-bit. I don't use distcc, either, since I'd rather not saturate my (802.11g) wireless network. -Os, in my experience, makes very little difference on amd64/x86_64/whate'er (FreeBSD 9.x amd64 clang the final sizes of the binaries between -Os & -O3 have very little bearing on expectations), however the atom seems to benefit quite a lot more from the smaller binaries, which on 32-bit (again in my experience with gentoo gcc46 i686) are significantly smaller than -O2 & -O3 binaries. I would assume this has to do with cache fit. Some suggest using -mfpmath=sse, which I've not studied in depth. But if the x87 bits are particularly slow on the atom (mind you, I have no idea), & you're not running stuff that makes heavy use of your sse registers I don't see a downside. 64-bit probably won't help much at all, unless you're running really numbery stuff. I suggest unless you're doing video editing or scientific number crunching (on your atom netbook) 32 will be fine.
Re: [gentoo-user] Installing specific layman packages & eix-test-obsolete
>> in package.mask: >> >> */*::init6 >> >> eix-test-obsolete find over 27,000 packages under this heading: >> >> Redundant in /etc/portage/package.mask: >> ... considered as REDUNDANT_IF_MASK_NO_CHANGE > > > The reason for this is the following: > Since the category and package is */*, your mask can match every > package - whether it matches depends on the package versions and > where they are from. So, for every package the following happens: > > eix -T checks for all versions of that package whether your > mask matches. Of course, for all packages which do not have a version > in the init6-overlay, the mask does not match, i.e., eix -T recognizes > that your mask has no effect for the tested package and thus outputs > the packge. Only packags which have a version in the init6-overlay > are not output. > > >> Adding the following to package.nowarn doesn't seem to change anything: >> >> */*::init6 mask_no_change > > > This does not help, since the packages which have no version in init6 > will not match this. > > You would need something like the opposite: > */*::gentoo mask_no_change > (and the same for all other overlays you are using). > However, this will suppress also warnings coming from other lines > in your package.mask file. > > So probably it is best if you disable the mask-no-change test > completely: The quickest is to put REDUNDANT_IF_MASK_NO_CHANGE=false > into /etc/eixrc. I also get 376 matches from "Not installed but in /etc/portage/package.mask" which are surely the packages in my overlays masked by */* but not installed. Do you know the name of this test so I can disable it in eixrc? I looked through man eix but couldn't find it. So REDUNDANT_IF_MASK_NO_CHANGE and the above test are not compatible with masking all but certain packages from an overlay via package.mask/package.unmask. I think what we really need is a better way to install only certain packages from an overlay. - Grant
Re: [gentoo-user] eix and bad colors.
On Fri, Dec 7, 2012 at 3:01 AM, Dale wrote: > Howdy, > > I was using eix a bit ago and I noticed the colors have changed. Since > I like to have white text and a black background, this is not working to > well for me. It seems some of the output is black text. Put black text > on a black background and I have missing text, usually the very thing I > am looking for. I have looked for a config somewhere in /etc but can't > find where this is set. I found where other colors are set but not for > eix. > > Anyone have a hint as to where this is set or is it hard coded into eix? > > Thanks much. > > Dale I think this may be fixed now. I no longer require the proposed edits to eixrc to get colors on a dark background. - Mark
Re: [gentoo-user] Installing specific layman packages & eix-test-obsolete
>> > This really freaks out eix-test-obsolete. Does anyone know of a way to >> > install only certain packages from a layman overlay and >> > use eix-test-obsolete? >> >> I don't add such overlays to make conf. Instead, I symlink directories for >> the packages I want into my local overlay. >> > > I did that, too. But it has become more difficult now. E.g. the 'progress' > overlay uses new EAPIs and new eclasses which even pull in incompatible packages, e.g. blas. > > I haven't found a way to handle that by symlinks of similar methods. I'm trying to see if I can get away with the symlink method. I removed the "source /var/lib/layman/make.conf" line from make.conf, created /usr/local/portage/sys-kernel/, and created a symlink in that directory like "geek-sources -> /var/lib/layman/init6/sys-kernel/geek-sources" but I get: * Missing digest for '/usr/local/portage/sys-kernel/geek-sources/geek-sources-3.7.0.ebuild' * Missing digest for '/usr/local/portage/sys-kernel/geek-sources/geek-sources-3.6.10.ebuild' * Missing digest for '/usr/local/portage/sys-kernel/geek-sources/geek-sources-3.5.7.ebuild' - Grant
Re: [gentoo-user] Intel Atom: architecture, distcc, crossdev and compile flags
Am 11.12.2012 18:36, schrieb Frank Steinmetzger: > Hello list > > Long time no read... :) > > It follows a verbose preamble. For the actual questions see dashed line > below. > TL;DR summary: it’s all about ricer-performance questions on a netbook. > > > I have the luck of having obtained a used netbook for free (Atom N450, single- > core with HT, 1 GB memory, 5400 rev HDD). During the last week I’ve been > experimenting with 32 and 64 bits on it and am still quite undecided which to > keep. My reasons: > > - They are not as far apart in CPU performance as is the Core2. > I posted a 32/64 comparison for Core2 a few months ago, which showed that > Lilypond speedup on 64 bit was 50%. On the Atom, it actually took 5% longer. > (Sadly, Blender doesn’t build on 32 bit right now). > - Startup times for hogs like Firefox and KDE are quite equal between the two > (that could be attributed in parts to the fact that the 64 bit partition > sits on the disk’s first sectors, while 32 bit sits at the other end, I > don’t know which end is faster). The first part (which actually is the outer edge of the physical disk). > - pro 64: it is very easy to use distcc, as opposed to 32 bits (see below). > - con 64: it uses about 50% more memory, 32 bit builds are a little faster. > > The RAM argument is the most convincing one right now, since more free RAM > means more cache, which means a faster system in the long run. Currently, KDE > after logon needs 150 MB on 32 bit, and 250 MB on 64 bit (without akonadi for > now). But awesome WM rocks on a netbook anyway. > > > --[ Questions begin ] > > So I’m interested in you opinion and own experience about the following > arising questions: > > * From my observations, the benefit of 64 bit over 32 is much smaller for an > Atom than it is for my Core2. Am I right to assume thus that the Atom > architecture doesn’t have much to offer to 64 bit (such as extra registers)? > I’m not talking about memory here, since it’s limited to 2 GB in any case. > It has the same set of registers as your Core2. It's just that the Atom micro-architecture is terrible with regard to 64bit. That's just about the only reason that x32 was invented (and now that I've said it, I'm just waiting for the flamewar about it). > * The problem of distcc between different architectures: > The netbook already had an older 32 bit Gentoo installed. And since I have > a multilib host (march=core2), I though I could upgrade with distcc (using > march=atom on the netbook). But at some point more and more stuff stopped > working, eventually I got “invalid instruction” errors during emerge, hence > I figured that was a dead end. > > So is it possible to mix architectures in this way at all with distcc? > I also have crossdev for i686 installed, which even shares files with the > system’s normal multilib gcc. I find that odd. I don't think you can mix x86_32 and x64 easily but I've never tried. Did you try adding a CFLAGS="-m32"? From comparing `gcc -Q --help=target -march=xxx` they should be compatible. > I sped up the installation process for 32 bit by using a chroot on the big > machine, which worked nicely. But it’s not a long-term solution, b/c it > uses up too much disk space on the host. > I do the same using NFS, bind mounts and tmpfs. What do you mean by disk space? If you can use common CFLAGS, you could try installing binary packages from your build host on your netbook (use quickpkg and friends). > * I’m interested in the question of -O2 vs. -Os. > Some sources say -Os is bad, b/c it breaks debugging and is mainly untested. > I won’t do heavy developing on it anyway, and Atoms do have a puny cache. > So I wonder whether -Os would improve execution time and RAM usage > noticably. Diskspace itself is not an issue. > I use -Os but have no data to back it up (I'm also using an older Atom). Gentoo wiki suggests using "-O2 -fno-reorder-blocks -fno-reorder-functions" as a compromise. > * I’m also interested in comparing bin packages over self-compiled ones. > E.g. I did compile icedtea, even if it’s just for TV browser. :) > Can you name a Java benchmark to measure CPU performance? > How about something from this site: http://shootout.alioth.debian.org/ > * The last thing I’m going to set up is filesystem encryption, at least for ~. > I already know/think that AES would be the best choice due to limited CPU > power, but what else is there to heed besides key size? > Nothing, you're good. Hash and key chaining method have negligible impact. If you stick with an x86_32 userspace I suggest at least using an x64 kernel so you can use of CRYPTO_AES_X86_64. > * What other small benchmarks for CPU and memory can you recommend? So far I > tested with nbench and sysbench. The results are so-and-so. Some computation > stuff is much slower on 64 bit, some a bit faster. The applic
Re: [gentoo-user] Intel Atom: architecture, distcc, crossdev and compile flags
I have: Atom(TM) CPU 330 @ 1.60GHz
running it with:
CFLAGS="-march=core2 -O2 -pipe"
CXXFLAGS="${CFLAGS}"
MAKEOPTS="-j5"
CHOST="x86_64-pc-linux-gnu"
Didn't run any test performance on it, but the only disappointing fact about
this small box is that it will not reboot itself when the power goes down.
It runs my asterisk and hylafax in remote location so I have to keep an eye on
it :-/
--
Joseph
On 12/11/12 18:36, Frank Steinmetzger wrote:
Hello list
Long time no read... :)
It follows a verbose preamble. For the actual questions see dashed line below.
TL;DR summary: it’s all about ricer-performance questions on a netbook.
I have the luck of having obtained a used netbook for free (Atom N450, single-
core with HT, 1 GB memory, 5400 rev HDD). During the last week I’ve been
experimenting with 32 and 64 bits on it and am still quite undecided which to
keep. My reasons:
- They are not as far apart in CPU performance as is the Core2.
I posted a 32/64 comparison for Core2 a few months ago, which showed that
Lilypond speedup on 64 bit was 50%. On the Atom, it actually took 5% longer.
(Sadly, Blender doesn’t build on 32 bit right now).
- Startup times for hogs like Firefox and KDE are quite equal between the two
(that could be attributed in parts to the fact that the 64 bit partition
sits on the disk’s first sectors, while 32 bit sits at the other end, I
don’t know which end is faster).
- pro 64: it is very easy to use distcc, as opposed to 32 bits (see below).
- con 64: it uses about 50% more memory, 32 bit builds are a little faster.
The RAM argument is the most convincing one right now, since more free RAM
means more cache, which means a faster system in the long run. Currently, KDE
after logon needs 150 MB on 32 bit, and 250 MB on 64 bit (without akonadi for
now). But awesome WM rocks on a netbook anyway.
--[ Questions begin ]
So I’m interested in you opinion and own experience about the following
arising questions:
* From my observations, the benefit of 64 bit over 32 is much smaller for an
Atom than it is for my Core2. Am I right to assume thus that the Atom
architecture doesn’t have much to offer to 64 bit (such as extra registers)?
I’m not talking about memory here, since it’s limited to 2 GB in any case.
* The problem of distcc between different architectures:
The netbook already had an older 32 bit Gentoo installed. And since I have
a multilib host (march=core2), I though I could upgrade with distcc (using
march=atom on the netbook). But at some point more and more stuff stopped
working, eventually I got “invalid instruction” errors during emerge, hence
I figured that was a dead end.
So is it possible to mix architectures in this way at all with distcc?
I also have crossdev for i686 installed, which even shares files with the
system’s normal multilib gcc. I find that odd.
I sped up the installation process for 32 bit by using a chroot on the big
machine, which worked nicely. But it’s not a long-term solution, b/c it
uses up too much disk space on the host.
* I’m interested in the question of -O2 vs. -Os.
Some sources say -Os is bad, b/c it breaks debugging and is mainly untested.
I won’t do heavy developing on it anyway, and Atoms do have a puny cache.
So I wonder whether -Os would improve execution time and RAM usage
noticably. Diskspace itself is not an issue.
* I’m also interested in comparing bin packages over self-compiled ones.
E.g. I did compile icedtea, even if it’s just for TV browser. :)
Can you name a Java benchmark to measure CPU performance?
* The last thing I’m going to set up is filesystem encryption, at least for ~.
I already know/think that AES would be the best choice due to limited CPU
power, but what else is there to heed besides key size?
* What other small benchmarks for CPU and memory can you recommend? So far I
tested with nbench and sysbench. The results are so-and-so. Some computation
stuff is much slower on 64 bit, some a bit faster. The applicability to
every-day use is of course a wibbly-wobbly argument.
I also tested the runtime of some application (packing and unpacking of
archives, throughput with dd, mencoder). If there is interest, I can post
the result of 21 runs on each platform, measured with GNU time.
--[ Questions end ]--
PS.: I’m aware that benchmarks are always a bit subjective and none is
perfect. I also realise that most of the questions quite belong into the
ricer corner. But Netbooks are ricer devices, b/c they need to perform at
their limits all the time. :-D
If you read until this point, thank you very much for your time. You get a
unicorn as a reward: `^nn~
--
Gruß | Greetings | Qapla’
Please do not share anything from, with or about me with any Facebook service.
“Time is money” said the waiter and put the date on the bill.
Re: [gentoo-user] Installing specific layman packages & eix-test-obsolete
I also get 376 matches from "Not installed but in /etc/portage/package.mask" which are surely the packages in my overlays masked by */* but not installed. Do you know the name of this test so I can disable it in eixrc? REDUNDANT_IF_IN_MASK (or in /etc/portage/package.nowarn: in_mask) I think what we really need is a better way to install only certain packages from an overlay. This is against the idea of an overlay: If you want only cerain packages copy them into your local overlay and do not add the whole overlay to portage. (But you might get troubles if you do not use eclasses or other ebuilds from the overlay which might contain corresponding patches). With eix you can still index such non-used overlays by using scripts like eix-layman (or more manually by setting ADD_OVERLAY in /etc/eixrc). For recent portage versions there was also a possibility to set global overlay order (IIRC by /etc/portage/overlay.conf or something similar; I do not remember since I never needed it).
[gentoo-user] crontab questions
Is there a way to remove "Cron " from the subject line of crontab mail without piping each cron job to 'mail'? I set 'usermod -c hostname root' on each of my systems so that the From: line displays "hostname" for crontab mail. This works on each system except the mail server itself which still shows "Cron Daemon". Can crontab mail from the mail server be made to display From: "hostname" like the other systems? I'm not completely clear on how cronbase works. Can this crontab be integrated into the system crontab via cronbase or should it be run as a separate user crontab for root? 0 4 * * * layman -NS && eix-sync -n && eix-remote update -n 15 4 * * * emerge -pvDuN world 20 4 * * * eclean -C distfiles 30 4 * * * eclean -C packages 40 4 * * * eix-test-obsolete 45 4 * * * revdep-rebuild -ip - Grant
Re: [gentoo-user] Installing specific layman packages & eix-test-obsolete
>> I think what we really need is a better way >> to install only certain packages from an overlay. > > This is against the idea of an overlay: If you want only cerain packages > copy them into your local overlay and do not add the whole overlay > to portage. (But you might get troubles if you do not use eclasses > or other ebuilds from the overlay which might contain corresponding > patches). Surely you can see the value in finding an ebuild for a package you like that isn't in portage, adding the overlay associated with that ebuild via layman, somehow specifying that you only want that package from the overlay, and running 'layman -S' to stay on top of version bumps. If I understand your suggestion correctly, it would involve manually checking for version bumps and recopying them into the local overlay. That doesn't seem very Gentoo. - Grant
[gentoo-user] recovery of fstab
Well I have a gentoo system I'm trying to recover. I've got it booted up via systemrescue. I do not have a copy of the fstab, so what is the best way to discover which partitions are /boot / and so on? (brain dead tonight) I guesses but the / is blank? df /dev/sda2 61438696 51276944 10161752 84% /mnt/gentoo/boot /dev/sda3 61438696 51276944 10161752 84% /mnt/gentoo /mnt/gentoo/boot is populated (mounted correctly) but the /dev/sda3 which I'm guessing is / is empty ? Been a while since I had to recover a system so referals to good docs are most welcome No to mention an automount capability with systemrescue? James
Re: [gentoo-user] Installing specific layman packages & eix-test-obsolete
On Tue, 11 Dec 2012 14:32:01 -0800, Grant wrote: > > This is against the idea of an overlay: If you want only cerain > > packages copy them into your local overlay and do not add the whole > > overlay to portage. (But you might get troubles if you do not use > > eclasses or other ebuilds from the overlay which might contain > > corresponding patches). > > Surely you can see the value in finding an ebuild for a package you like > that isn't in portage, adding the overlay associated with that ebuild > via layman, somehow specifying that you only want that package from the > overlay, and running 'layman -S' to stay on top of version bumps. If I > understand your suggestion correctly, it would involve manually checking > for version bumps and recopying them into the local overlay. That > doesn't seem very Gentoo. Agreed. I may not want to add the whole overlay. That's fine for things like the kde or vmware overlays that have a specific focus, but sometimes I find a package I want isn't in portage but is in an overlay, but adding that overlay pulls in all sorts of other packages, sometimes beta or unstable versions. However, the idea of setting overlay order sounds interesting as it would at least solve part of the problem. For now, symlinking rather than copying, which I tried before, at least keeps things up to date. -- Neil Bothwick Snacktrek, n.: The peculiar habit, when searching for a snack, of constantly returning to the refrigerator in hopes that something new will have materialized. signature.asc Description: PGP signature
Re: [gentoo-user] recovery of fstab
mount them and see whats there? also, what order did you mount them in? it may make a difference On Tue, Dec 11, 2012 at 2:48 PM, James wrote: > Well I have a gentoo system I'm trying to recover. > I've got it booted up via systemrescue. > > I do not have a copy of the fstab, so what is the best > way to discover which partitions are /boot / and so on? > > (brain dead tonight) > > > I guesses but the / is blank? > > df > /dev/sda2 61438696 51276944 10161752 84% /mnt/gentoo/boot > /dev/sda3 61438696 51276944 10161752 84% /mnt/gentoo > > > /mnt/gentoo/boot is populated (mounted correctly) but the /dev/sda3 > which I'm guessing is / is empty ? > > Been a while since I had to recover a system so referals to good > docs are most welcome > > No to mention an automount capability with systemrescue? > > James > > > > > >
Re: [gentoo-user] recovery of fstab
On Tue, Dec 11, 2012 at 10:48:03PM +, James wrote: > > I guesses but the / is blank? > > df > /dev/sda2 61438696 51276944 10161752 84% /mnt/gentoo/boot > /dev/sda3 61438696 51276944 10161752 84% /mnt/gentoo > > > /mnt/gentoo/boot is populated (mounted correctly) but the /dev/sda3 > which I'm guessing is / is empty ? > > Been a while since I had to recover a system so referals to good > docs are most welcome > > No to mention an automount capability with systemrescue? Start by issuing "blkid" to determine what partitions are there. If you use -L with mkfs. then you should have a label, such as this: peter ~ # blkid /dev/sda1: LABEL="root" UUID="73362905-79dc-4512-9518-4c040963f80e" TYPE="xfs" /dev/sda2: LABEL="boot" UUID="2559cf58-63fa-44d3-ac82-94514d4d3769" TYPE="xfs" /dev/sda3: LABEL="home" UUID="8d5d17c1-7f99-4a49-b256-4d075306e42e" TYPE="xfs" /dev/sda5: LABEL="swap" UUID="636e7622-4a10-43d2-b976-d37cc15da7fc" TYPE="swap" Then "mkdir /tmp/help" or whatever, and "mount /dev/ /tmp/help/ and when you get the old / located, mount it first later. mount /dev/ /mnt/gentoo mount /dev/sda2/mnt/gentoo/boot It looks like sda2 and sda3 are identical, so you got something mounted wrong. Bruce -- Happy Penguin Computers >') 126 Fenco Drive ( \ Tupelo, MS 38801 ^^ supp...@happypenguincomputers.com 662-269-2706 662-205-6424 http://happypenguincomputers.com/ Don't top-post: http://en.wikipedia.org/wiki/Top_post#Top-posting
Re: [gentoo-user] recovery of fstab
> mount them and see whats there? > Also, check the partition table to see which one is swap, and what other partitions exist.
Re: [gentoo-user] Installing specific layman packages & eix-test-obsolete
> > > This is against the idea of an overlay: If you want only cerain
> > > packages copy them into your local overlay and do not add the whole
> > > overlay to portage. (But you might get troubles if you do not use
> > > eclasses or other ebuilds from the overlay which might contain
> > > corresponding patches).
> >
> > Surely you can see the value in finding an ebuild for a package you like
> > that isn't in portage, adding the overlay associated with that ebuild
> > via layman, somehow specifying that you only want that package from the
> > overlay, and running 'layman -S' to stay on top of version bumps. If I
> > understand your suggestion correctly, it would involve manually checking
> > for version bumps and recopying them into the local overlay. That
> > doesn't seem very Gentoo.
>
> Agreed. I may not want to add the whole overlay. That's fine for things
> like the kde or vmware overlays that have a specific focus, but sometimes
> I find a package I want isn't in portage but is in an overlay, but adding
> that overlay pulls in all sorts of other packages, sometimes beta or
> unstable versions.
>
> However, the idea of setting overlay order sounds interesting as it would
> at least solve part of the problem. For now, symlinking rather than
> copying, which I tried before, at least keeps things up to date.
According to Helmut, symlinking isn't working as well as it used to due to new
EAPIs and new eclasses. I'm getting missing digest errors when I try it:
* Missing digest for
'/usr/local/portage/sys-kernel/geek-sources/geek-sources-3.7.0.ebuild'
* Missing digest for
'/usr/local/portage/sys-kernel/geek-sources/geek-sources-3.6.10.ebuild'
* Missing digest for
'/usr/local/portage/sys-kernel/geek-sources/geek-sources-3.5.7.ebuild'
I'm also getting this from eix-test-obsolete and not sure why:
Installed packages with a version not in the database (or masked):
[D] sys-kernel/geek-sources (3.6.10(3.6.10){tbz2}[2]@12/11/12 ->
~3.4.19(3.4.19)^mbs[1] ~3.4.21(3.4.21)^mbs[2] ~3.5.7(3.5.7)^mbs[2]
~3.6.8(3.6.8)^mbs{tbz2}[2] ~3.6.9(3.6.9)^mbs{tbz2}[2])
I've run 'eix-remote update' and I
have /var/lib/layman/init6/sys-kernel/geek-sources/geek-sources-3.6.10.ebuild.
Is the warning triggered from */*::init6 in package.mask even though I
have sys-kernel/geek-sources::init6 in package.unmask? If so, which test
should I disable for this one?
- Grant
Re: [gentoo-user] Intel Atom: architecture, distcc, crossdev and compile flags
On Tue, Dec 11, 2012 at 09:20:55PM +0100, Florian Philipp wrote: > > * From my observations, the benefit of 64 bit over 32 is much smaller for an > > Atom than it is for my Core2. Am I right to assume thus that the Atom > > architecture doesn’t have much to offer to 64 bit (such as extra > > registers)? > > I’m not talking about memory here, since it’s limited to 2 GB in any case. > > > > It has the same set of registers as your Core2. Incidentally, when I initially set up the netbook, the output of gcc -march=native -E -v - &1 | sed -n 's/.* -v - //p' (which floated around the ML in the past) implied core2, IIRC. > It's just that the Atom micro-architecture is terrible with regard to > 64bit. That's just about the only reason that x32 was invented (and > now that I've said it, I'm just waiting for the flamewar about it). Terrible in what way? Inadequate memory throughput? I didn't know x32, but from what I've read in the last few minutes it sounds intriguing. > > So is it possible to mix architectures in this way at all with distcc? > > I also have crossdev for i686 installed, which even shares files with the > > system’s normal multilib gcc. I find that odd. > > I don't think you can mix x86_32 and x64 easily but I've never tried. > Did you try adding a CFLAGS="-m32"? From comparing `gcc -Q --help=target > -march=xxx` they should be compatible. Not yet, putting in on todo, as it would involve building, running, testing (or reading up on it :-p ). > > I sped up the installation process for 32 bit by using a chroot on the big > > machine, which worked nicely. But it’s not a long-term solution, b/c it > > uses up too much disk space on the host. > > I do the same using NFS, bind mounts and tmpfs. What do you mean by disk > space? That I don't have much space left on the host machine for the entire chroot. I bind-mount distfiles and portage, but I'm still running low on gigabytes. I was thinking of NFS quite early, but a friend said it would perform not nicely. Also, with all my cables currently occupied, the two are connected over a slow WiFi router. It's one of the rare cases in which compressing distcc traffic increases performance. :) The netbook has gigabit ethernet, though. Thank $DEITY for compressed tar pipes over SSH. I wonder what Windows people would do in such a situation. >:-] > If you can use common CFLAGS, you could try installing binary > packages from your build host on your netbook (use quickpkg and friends). Which brings me back to the disk space problem. Right now the 32 bit chroot is a direct mirror, which is rsynced to and fro when the netbook runs in 64 bit. Once I have a cable available I think I'll go with NFS. > > […] Can you name a Java benchmark to measure CPU performance? > > How about something from this site: > http://shootout.alioth.debian.org/ Will have a look-see. > > > * The last thing I’m going to set up is filesystem encryption, at least for > > ~. > > I already know/think that AES would be the best choice due to limited CPU > > power, but what else is there to heed besides key size? > > Nothing, you're good. Hash and key chaining method have negligible > impact. If you stick with an x86_32 userspace I suggest at least using > an x64 kernel so you can use of CRYPTO_AES_X86_64. That's an interesting idea. If I had a 32 bit userland, I would have to build new kernels on my big 64 laptop then, right? I don’t suppose I can simply mix chosts, such that I would have a multilib x86_64 gcc/binutils/glibc, but i686 everything else. I haven't done any comparisons of 32/64 crypto yet, I'm just reading docs on Luks (never used it before). Big stuff (videos, music) won't be encrypted anyway, just the sensitive data like mail, documents, passwords and personal photos. So the requirements won't be high. However I might expand it to /, though that would involve a more complicated boot process (I never needed initrds except for bootsplash). On a sidenote, While I was cleaning up unread mails in the ML, I just found your interesting frontswap/zcache trick. I wonder how many years I'd have to use the device to get back the time from improved performance that I spent setting it up in the first place. :-D > > * What other small benchmarks for CPU and memory can you recommend? > > […] > > How about trying some browser benchmarks. […] I could also use those to compare binary and source firefox (which is compiling in the chroot right now). > There is also a Qt render benchmark > http://code.google.com/p/qtperf/ > Check out app-admin/eselect-qtgraphicssystem and see how they compare in > appearance and numbers. Nice idea, since I'm a general Qt fanboy. So thanks a lot for the info so far, I'll try to digest it all tomorrow. Good night for now. *yawn* -- Gruß | Greetings | Qapla’ Please do not share anything from, with or about me with any Facebook service. Emacs is a great operating system, which only lacks a good editor. pgp1
[gentoo-user] Re: recovery of fstab
Bruce Hill happypenguincomputers.com> writes: > Start by issuing "blkid" to determine what partitions are there. If you use -L > with mkfs. then you should have a label, such as this: > peter ~ # blkid > /dev/sda1: LABEL="root" UUID="73362905-79dc-4512-9518-4c040963f80e" > TYPE="xfs" I used parted -l > mount /dev/≤old /> /mnt/gentoo > mount /dev/sda2/mnt/gentoo/boot I got it mounted now: /dev/sda2 208820 80584128236 39% /mnt/gentoo/boot /dev/sda3 61438696 51276944 10161752 84% /mnt/gentoo/new thx James
[gentoo-user] udev: boot failure
OK so lots of updates (mostly kde 4.9.3) to several systems today. 2 or the 3 are fine One got hosed. Error message upon reboot config_devdmpfs=y is required in your kernel configuration for this version of udev to run successfully this requires immediate action mdev: sys/class no such file mdev Error udev-mount failed to start Error cannot start udev as udev-mo9unt would not start So I booted the system with system rescue and got the partitions mounted (boot and /) / contains /var and /usr looking into : /mnt/gentoo/new/usr/portage/sys-fs/udev I cannot tell what version of udev (udev mount) and others (udev-init-scipts ?) were installed and how to roll this back. Any guidance on rolling back these errors is most welcome James
[gentoo-user] Re: udev: boot failure
James tampabay.rr.com> writes: > looking into : /mnt/gentoo/new/usr/portage/sys-fs/udev > I cannot tell what version of udev (udev mount) > and others (udev-init-scipts ?) were installed > and how to roll this back. OK so on one of my working systems, I have sys-fs/udev-171-r9 installed. So I should go chroot and downgrade to udev-171-r9? Any other packages to roll back? Googling did not produce anything useful. James
Re: [gentoo-user] udev: boot failure
On Wednesday 12 December 2012 08:07:46 AM IST, James wrote: > > OK so lots of updates (mostly kde 4.9.3) > to several systems today. 2 or the 3 are fine > > One got hosed. Error message upon reboot > > config_devdmpfs=y is required in your kernel configuration > for this version of udev to run successfully > this requires immediate action > mdev: sys/class no such file mdev > Error udev-mount failed to start > > Error cannot start udev as udev-mo9unt would not start > > So I booted the system with system rescue > and got the partitions mounted > (boot and /) / contains /var and /usr > > looking into : /mnt/gentoo/new/usr/portage/sys-fs/udev > I cannot tell what version of udev (udev mount) > and others (udev-init-scipts ?) were installed > and how to roll this back. > > Any guidance on rolling back these errors is most > welcome > > James > > > > It's not a udev problem. You need to recompile your kernel with devtmpfs support. It can be found in device-drviers -> generic driver options. -- Nilesh Govindarajan http://nileshgr.com
Re: [gentoo-user] Re: udev: boot failure
On Wednesday 12 December 2012 08:19 AM, James wrote: > James tampabay.rr.com> writes: > > >> looking into : /mnt/gentoo/new/usr/portage/sys-fs/udev >> I cannot tell what version of udev (udev mount) >> and others (udev-init-scipts ?) were installed >> and how to roll this back. > > OK so on one of my working systems, I have > sys-fs/udev-171-r9 installed. So I should > go chroot and downgrade to udev-171-r9? > > Any other packages to roll back? > Googling did not produce anything useful. > > > James > > Downgrading udev would probably put you in a much bigger soup. Just recompile your kernel with devtmpfs support. You might want to use a recovery disk sort of thing or some live usb/cd. -- Nilesh Govindarajan http://nileshgr.com
Re: [gentoo-user] crontab questions
On 12/11/2012 04:15 PM, Grant wrote: > Is there a way to remove "Cron " from the subject line of > crontab mail without piping each cron job to 'mail'? > > I set 'usermod -c hostname root' on each of my systems so that the From: > line displays "hostname" for crontab mail. This works on each system > except the mail server itself which still shows "Cron Daemon". Can > crontab mail from the mail server be made to display From: "hostname" > like the other systems? > > I'm not completely clear on how cronbase works. Can this crontab be > integrated into the system crontab via cronbase or should it be run as a > separate user crontab for root? > > 0 4 * * * layman -NS && eix-sync -n && eix-remote update -n > 15 4 * * * emerge -pvDuN world > 20 4 * * * eclean -C distfiles > 30 4 * * * eclean -C packages > 40 4 * * * eix-test-obsolete > 45 4 * * * revdep-rebuild -ip > If your goal is to run these each one after the other, you can simply stick a shell script in /etc/cron.daily that executes them in order. The default crontab runs any executable files in, * /etc/cron.daily * /etc/cron.hourly * /etc/cron.monthly * /etc/cron.weekly at roughly the time specified in /etc/crontab. If any of those directories contain scripts, they're run in "alphabetical" order, i.e. how `ls` would sort them. To fix the Subject/From headers, try, http://www.postfix.org/header_checks.5.html I've never had to use them myself, but I think the REPLACE action will do what you want. The alternative is to replace the sendmail binary with something that executes e.g., sed -e 's/Subject: Cron <[^>]> /Subject: /g' | /the/actual/sendmail Both feel a little dirty, but the header checks are less likely to break something assuming that they will work on a client-provided From header.
[gentoo-user] Mounting floppy disks
For some reason, when I mount floppy disk (standard HD 3.5" VFAT disk) it does nothing. No error, just nothing... I have not tried this in well over a year, but it used to work. The /dev/fd0 device works normally, I can access it with mtools and use dd and even access disks in virtual machines running mswindows. Furthermore, if I make an image of the disk using dd and then loop mount that image, it works! But when I try to mount my normal /mnt/floppy fstab entry that I've had for years, or manually "mount -t vfat /dev/fd0 /mnt/floppy" there is disk activity, but it never shows up in /proc/mounts and is not mounted for me afterward. If I add "-v" to mount, it tells me that it mounted, but it didn't actually... either that or it has unmounted itself instantly before I can see it. Can anyone else mount floppies? Is there some floppy magic I have forgotten over the years? Thanks for any tips!
Re: [gentoo-user] Localmount starts before LVM
On Tue, 11 Dec 2012 10:46:19 -0500 Randy Barlow wrote: > Alan McKinnon wrote: > > On Tue, 11 Dec 2012 18:28:37 +0530 > > Nilesh Govindrajan wrote: > >> Exactly the reason why I wanted RAID0 and LVM in combination: more > >> IOPS. ZFS looks very interesting, how stable is it? > > > > On Linux, not at all (it doesn't exist there except using fuse) > > > > On FreeBSD, rock solid. > > On Solaris, rock solid. > > > > It almost seems to be everything btrfs is not... > > The details why this is the case are something I can never remember > straight in my head, but I recall that it's due to licensing that ZFS > cannot be included in the Linux kernel directly. I think it might be > because the ZFS license doesn't have the Copyleft clause that the GPL > requires? That's the one - The ZFS license from Sun is incompatible with GPL-2 That only stops Linus and distros from redistributing the code, the rest of us are free to downloaded it, patch the kernel and run it to our heart's content. > > It's sad, because ZFS is really pretty great. I think btrfs will be > pretty great too once it is stabilized, so I look forward to that. > > Also, I had seen some kernel patches that you can apply yourself to > get ZFS in Linux without FUSE a year or two back. I never tried them, > and can't attest to how stable or unstable they might be, but you > could look into that as well. > -- Alan McKinnon alan.mckin...@gmail.com
Re: [gentoo-user] System maintenance procedure?
On Tue, 11 Dec 2012 07:36:10 -0600 Bruce Hill wrote: > On Sun, Dec 09, 2012 at 04:48:24PM +, Neil Bothwick wrote: > > > > > workstation ~ # emerge -a @preserved-rebuild > > > emerge: 'preserved-rebuild' is an empty set > > > emerge: no targets left after set expansion > > > > So you have nothing that needs rebuilding. Portage will warn you > > when the set it non-empty, telling you to run emerge > > @preserved-rebuild. There is no need to run it at any other time. > > After using Gentoo for close to two years, the only time/place I've > ever even seen @preserved-rebuild is in this thread. Yet you say, > "Portage will warn you when the set is [it] non-empty, telling you to > run emerge @preserved-rebuild." > > How will portage do this? An alias 'ud' > > alias ud='eix-sync && emerge -aDjNuv @world && dispatch-conf && > emerge -a --depclean && revdep-rebuild -i && clear && exit' > > is run every morning with my first cup of coffee. If something were > changed or left off that alias do you suppose this mysterious > @preserved-rebuild would be run? No, you would likely never see it. Your alias runs revdep-rebuild, which would inelegantly fix the very problem that @preserved-rebuild elegantly fixes. Of course, all this assumes that your version of portage supports @preserved-rebuild To use it, you simply notice the portage message right at the end of an emerge and run "emerge @preserved-rebuild" - it's just a regular emerge using a particular built-in set that has a defined purpose -- Alan McKinnon alan.mckin...@gmail.com
