Re: [gentoo-user] Boot disk renames /dev/md6 to /dev/md127.
I came across this too. The misnamed device is still perfectly usable. If you boot from something with an unconfigured or misconfigured initramfs (such as the install CD?) and need to chroot into a system, deactivating the array and reassembling it, even using --scan, should name it properly. Configuring /etc/mdadm.conf and then setting MDADM_CONFIG in /etc/genkernel.conf (or supplying --mdadm-config) before creating your initramfs will name the device properly. If you are not using an initramfs then I'm not sure this will be a persistent issue.
Re: [gentoo-user] Boot disk renames /dev/md6 to /dev/md127.
Of course, there might be other causes, but if it just happened randomly I suspect the above is the most likely.
Re: [gentoo-user] Digital Cinema 4k (4096x2160@60Hz) SST mode display and xf86-video-ati
I don't think anybody is capable of giving you a firm answer because nobody has done it. You would probably be the first. In fact, when double-checking the info I had found earlier, some of the searches gave your message in the first 3-4 results. Hi Mom! Cards which support UHD should support 4k. You can try to use the closed driver benchmarks from Phoronix to guesstimate if your intended usage will be passable (http://www.phoronix.com/scan.php?page=articleitem=ubuntu_1404_4kgpus). The R7 250E may have issues running a composited desktop. I would not expect the R9 280X to have issues. Most complaints seem to be from people with older cards, but then again, those reporting success all seem to be using closed drivers. I rated success likely enough that I was considering a Dell P2415Q w/ the Intel HD4600 (and replacing my motherboard) or an R9 290X. Make sure you have an exit strategy, though.
Re: [gentoo-user] android development with Qt in a Gentoo box
A bit ago I was surprised to find out that http://doc.qt.io/qt-5/android-support.html exists. You might look at it, I can't personally recommend it. See also http://doc.qt.io/qt-5/platform-notes-android.html. I tend to suggest people at least look at the Google provided workflow. There are issues if you attempt to deviate from it which inevitably go back to the design of Android. Your device wasn't meant to do a lot, sadly.
Re: [gentoo-user] Calculating dependencies...: Any way to make it faster?
If the bottleneck is reading the information from disk you might upgrade the SD card or use a USB drive instead, which may have better random access performance. You could also store the portage tree on another machine with faster storage and access it over the network. If the bottleneck is actually calculating the dependencies, you are probably out of luck for the immediate future. For calculating dependencies I suspect the larger bottleneck is reading everything from disk, seeing as the machines seem fast enough you didn't complain too much about actually compiling. In either case you should try to revisit distcc and cross compiling as that is the only reliable way to speed everything up. You do not necessarily need to use distcc with a cross compiler (the configuration most likely to cause problems, though the wiki does address this).
Re: [gentoo-user] Re: Emergeing Firefox 35.0 breaks Ghostery ... fixable?
I might suggest Adblock Edge as an alternative. It seems to work. Also, if you choose an EasyList+EasyPrivacy subscription, it should do all of what Ghostery does. An equivalent extension to replace Ghostery would be Disconnect.Me (it does not have ties to an advertising company, if you cared).
Re: [gentoo-user] Is it wrong to install a specific version
A package set is just a list of packages in a file under /etc/portage/sets. You can operate on every package in the set at once. https://dev.gentoo.org/~zmedico/portage/doc/ch02.html If you merged the packages by version they won't be automatically updated. This includes the case where they are part of a set (unless you leave part of the version unspecified). If you merged the slot, the packages would update to changes made in that slot.
Re: [gentoo-user] Firefox bookmarks
Look at the contents of ~/.mozilla/firefox and ~/.mozilla/firefox/profiles.ini in particular. The solution should be obvious. Also, see: https://support.mozilla.org/en-US/kb/profiles-where-firefox-stores-user-data#w_working-with-profiles, the first Google result. ಠ_ಠ
Re: [gentoo-user] NSA SELinux kernel support
...until it doesn't, and then what? The comment was slightly off-topic and mainly pointed towards his decision to disable SELinux on a distribution which had enabled it by default. On Gentoo, if you enable SELinux, see all of the AVCs and decide to nope right out of there, you are making an informed decision (by virtue of needing to learn a great deal about SELinux to set it up in the first place). I could have half-assed it with audit2allow, but security-wise that's a cop-out. I'm not sure it's a complete cop-out as long as you read the suggestions audit2allow is making. The policy you end up with will not be ideal and will certainly be full of holes, but at least you are somewhat aware of the risk a given service is to your system. I'd like to find a middle ground, and it might be Targeted mode (I was attempting Strict). Or, it might be a different system like AppArmor. Yeah, my ending suggestion was to run in targeted mode (if you wanted to bother with SELinux at all) but that mainly serves as a workaround for Desktop-oriented stuff. Containers or virtualization are also options.
Re: [gentoo-user] setxkbmap -option ctrl:nocaps
You should be using xmodmap OR xsetkbmap OR the configuration file, not all three (though they shouldn't interfere with each other if configured properly - they'd just all be doing the same thing). If you wish to set Caps_Lock to be Control_L, you need to redefine keycode 66, not 37 (37 is Control_L by default). You may also need to add clear lock or clear Lock before that line. Sometimes it will or will not work if you don't do this and will toggle Caps_Lock and press Control_L. Weird. Alternatively, you can use xsetkbmap like you described.
Re: [gentoo-user] NSA SELinux kernel support
I was wondering if there was any harm in disabling the NSA SELinux support in my gentoo-sources based kernel. There is no harm, but if you were interested a lot of packages come with policies by default. Currently there is no support for SELinux in Gentoo for the vast majority of desktop applications. It is a little bit of work to get anything nonfunctional working. There are additional modes where you can simply run your user as unconfined and any services will be restricted by SELinux. grsecurity's RBAC is an alternative where you simply let it generate a policy based on what it sees you use. Notably, Fedora and CentOS enable SELinux by default. SELinux is the only one I've had a bit of experience with - I run CentOS (SELinux is enabled by default) for some personal-use-only services that I want to run without dealing with Gentoo. My first step in a CentOS install is to disable SELinux (and the firewall, hehe) to avoid dealing with the pain of wading through documentation for hours on end. http://stopdisablingselinux.com/ - your distribution probably comes with policies for everything you want to install, anyway...
Re: [gentoo-user] VMs - what technology would you advise?
Yes, in favor of KVM. http://www.phoronix.com/scan.php?page=articleitem=intel_haswell_virtualization http://www.phoronix.com/scan.php?page=articleitem=ubuntu_1404_kvmboxt On Wed, Dec 31, 2014 at 5:44 AM, Mick michaelkintz...@gmail.com wrote: On Wednesday 31 Dec 2014 07:32:18 Sid S wrote: I would suggest QEMU/KVM takes the place of VirtualBox. I've not actually found anything it doesn't support, though VirtualBox is far more polished. Starting a VM will be as easy as running a shell script (or you can use virt-manager). Thanks Sid, other than the GUI and potential ease of use, is there a difference in performance between Vbox and KVM? -- Regards, Mick
Re: [gentoo-user] VMs - what technology would you advise?
On Wed, Dec 31, 2014 at 7:26 AM, Mick michaelkintz...@gmail.com wrote: Vbox seems to be coming last by quite some margin in the intel tests! I also read this article and it looks that vbox is thankfully doing better on AMD; but there are differences in the versions and kernels used between the two articles: Yes, but it didn't seem relevant to your usecase, so I didn't lead with it. I was just hoping you might consider it at some point as an alternative to VirtualBox, as it is fairly complete at this point (though sans memory snapshotting, which is a useful feature I had not considered - I had been doing pretty well with disk snapshots). I originally researched virtualization with an eye to making it usable on a laptop/notebook. In this regard VT-x/VT-d with KVM give you usable battery lifespan and let you use less powerful hardware. Implication: You might not need the workstation to do your testing, depending on what testing you do. Something to consider for the future. I was elated to find I did not need to tie myself to a beefy machine to do what little Windows/.NET development I indulge in. Containers and such definitely sound interesting; I had been avoiding Linux VMs for the longest time due to the overhead. The alternatives sound rather light so I might reconsider.
Re: [gentoo-user] VMs - what technology would you advise?
I would suggest QEMU/KVM takes the place of VirtualBox. I've not actually found anything it doesn't support, though VirtualBox is far more polished. Starting a VM will be as easy as running a shell script (or you can use virt-manager).
Re: [gentoo-user] pavucontrol - Pulseaudio Volume Control, save settings
Pulseaudio is started automatically by either your desktop environment or programs which make use of it. It's running under your user. To solve your problem, I would look at adding `pactl` or `pacmd` line to any startup scripts you might have available. Note I don't actually think you are experiencing pulseaudio forgetting your settings - I'm pretty sure the default behavior just prefers external devices to internal ones (so, you're fighting pulseaudio in this case).
Re: [gentoo-user] pavucontrol - Pulseaudio Volume Control, save settings
Sorry, I forgot to add you are probably going to want to set the default source.
Re: [gentoo-user] New PC, new boot concepts
I still use GRUB to boot ISO images, and have a sysrescd image in /boot for this. With UEFI, you can have more than one bootloader installed. I hadn't considered the possibility of UEFI booting to an ISO directly, I'd be interested to know if it is possible. Typically one takes the contents of the iso, places that on a partition, and then makes it bootable with grub. I'm not sure you can boot the iso directly. Finally, what's your opinion on 'secure boot'? That the main security is is aimed at is job security for MS employees. I turn it off straight away. It is not a completely stupid idea. When it is adequately supported it is something I intend to use. It is possible (unless I misunderstood something) to provide your own keys and sign your own OS partition. Anyway, I am kind of surprised people are still having problems with UEFI. It's necessary to turn SecureBoot off, but otherwise I just got everything to work.
Re: [gentoo-user] Nvidia and optimus
I'd like to use the internal card most of the time since I don't care about 3D acceleration but I do care alot about power saving. When using an external monitor I'd like to use the NVidia card. Currently my solution is to reboot and change bios settings, being able to switch at runtime would be a real enhancement for me. Is this possible? Yes, install xrandr/xinerama. P.S. On top of that, this laptop has only a year or so left before I replace it, and I know now to avoid Optimus entirely in the future. Will probably be impossible in the future if you buy anything with a discrete card (even a crap one, for multi-monitor). New laptops are shipping without an HDMI multiplexer like Christian's has.
Re: [gentoo-user] Identifying a file by a block number...how?
find / -xdev -inum # If you know it is in a directory more specific than /, replace / with that directory. inodes are only meaningful to ext2/3/4, but you can use the fs tools to find out where it is on disk. On Thu, Dec 18, 2014 at 3:14 PM, Poison BL. poiso...@gmail.com wrote: On Thu, Dec 18, 2014 at 3:24 PM, meino.cra...@gmx.de wrote: Hi, with sysctl vm.block_dump=1 one can enable the logging of IO to the harddisk/flashmem/... into dmesg. The logs report the block number of the file in question... but not the filename itsself. Is there any other way as examine each single file of the filesystem to find the file to which a certain block number is assigned? Thank you very mcuh for any help! Best regards, Meino That depends entirely on the filesystem being used. In the case of ext2/3/4, I believe /sbin/debugfs will do the trick with its icheck command to get the inode, and once you have the inode, you can get the filename via find. What I'm not 100% certain of is whether the block numbers involved map 1:1 with physical sectors, and how that plays with the 512B vs 4KB sectors, etc. With NTFS it's a hair quicker with ntfscluster and ntfsinfo doing the trick fairly trivially (I use a tool centered around that combo to identify files lost when I recover peoples windows drives with ddrescue). -- Poison [BLX] Joshua M. Murphy
Re: [gentoo-user] Nvidia and optimus
I got bumblebee to work on hardened w/ SELinux. It is definitely ready for use. Anyway, as has been explained, you basically HAVE to use optimus - if you would like to return your laptop and/or sue the laptop manufacturer for false advertising, now would be the time to do it. On Tue, Dec 16, 2014 at 3:39 PM, Heiko Baums li...@baums-on-web.de wrote: Am 16.12.2014 um 12:55 schrieb behrouz khosravi: I have not tried the bumblebee. You need bumblebee. Otherwise it's not possible to use the Nvidia Optimus chip. I just waned to use optimus without that, but it seem the it is not easy! It's not possible, because the Nvidia Optimus chip isn't a full featured graphics card, and doesn't write directly to the screen. Joost already explained it pretty well. The 2D graphics is done by the GPU embedded in the CPU, which also writes the output to the screen. The Nvidia Optimus chip is only a helper chip to do the additional 3D rendering. It gives its output to the GPU embedded in the CPU which in turn writes the output to the screen. To use the Nvidia Optimus chip you need to install these packages: x11-misc/bumblebee x11-misc/virtualgl sys-power/bbswitch x11-drivers/nvidia-drivers I don't know if, but I don't think that, it will work with x11-drivers/xf86-video-nouveau. Then you need to add bumblebee and vgl to your default runlevel. rc-update add bumblebee rc-update add vgl To run a 3D application you need to start it with `optirun command`. And don't try to `eselect opengl set nvidia`. This won't work for the described reasons. You need to `eselect opengl set xorg-x11`. I think I will try that sometime It's actually quite easy and the Nvidia Optimus support by bumblebee is pretty good. The reason why this is done this way is power saving. 3D rendering is pretty power-consuming. Heiko
Re: [gentoo-user] Nvidia and optimus
Using nouveau doesn't imply the card will always be on. Assuming the firmware on the device turns it off and on, vga-switcheroo or bbswitch will send the proper ACPI commands to turn it off and on. When it is on the chosen driver will be used. On Tue, Dec 16, 2014 at 5:46 PM, Erik Mackdanz erikm...@gmail.com wrote: Heiko Baums li...@baums-on-web.de writes: I don't know if, but I don't think that, it will work with x11-drivers/xf86-video-nouveau. I used bumblebee for quite a while. It worked okay, but every upgrade I would have to fiddle with it again. I switched to the Nouveau driver and I'm very glad I did. Conventional wisdom says Nouveau quality is lower than Nvidia, but I found it worked better on some things (Second Life). As someone else pointed out, with Nouveau the GPU remains on all the time consuming power. This is the downside. If ease-of-use and/or open licensing are more important to you than top rendering quality and power consumption, consider using Nouveau. -- Erik Mackdanz
Re: [gentoo-user] Re: Gentoo's future directtion ?
Ah, I apologize. I did not mean to quote anything (it happens automatically; I will start paying more attention to it). I was hoping he would remember his additions to the conversation at hand and could extrapolate on them.
Re: [gentoo-user] Re: Gentoo's future directtion ?
Hasufell, what are you referring to by attempts to make Gentoo more friendly to gaming? On Sat, Nov 29, 2014 at 1:38 PM, hasufell hasuf...@gentoo.org wrote: Alan Mackenzie: So that instead of conceptualising a branch (as you would do with Mercurial, Bazaar, Subversion, or even CVS), you need to think about commits reachable from a certain head (excluding commits reachable from some other head). [snipping everything that is not technical] How exactly is that a disadvantage? You are just complaining about the way a concept is presented without saying what actual limitations that implies (if any). If you like mercurial better, use that. Speaking about disadvantages however requires a bit more than I like that way better.
Re: [gentoo-user] Gentoo's future directtion ?
We didn't disband the team because we thought that having a team focused on games wasn't a bad idea, but so far nobody else seems all that interested so it seems as likely as not that there won't be a games team in the future. Probably a chicken-and-egg thing. I want to play games on my Gentoo, or a vm hosted by Gentoo, but doing so can be a pain. Is there any reason to cater to this demographic specifically? Well, no, but making it less hard would require solving some nontrivial problems others might benefit from. Which is the purpose of Gentoo, right? As for Java, I've not encountered any major bugs. This might be why nobody is talking about Java bugs. If there are any bugs with the introduction of Java 1.8, again, there's no reason to cater to the demographic that wants Java 8, but it will likely solve hard problems. Java is something I should be able to use by accident. It is almost entirely self-contained. Solving these hard problems is likely to help in other areas. For example, let me refer to another portion of the conversation below: The current Gentoo way is far more limiting, but by having a single version of glibc with a single policy around versioning/etc packages don't have to micromanage what they depend on. I actually think this is wrong. The Gentoo way might be limiting in some respects, but the reality seems to be it is limited by the software it is working with more than the other way around. So... what's a better way to do things? NI,SF. Do-autocracy suffers from the challenging problem of challenging problems being avoided. On top of that, this would have to be an issue that has to be handled by the software devs. If only the universe were ebuilds and not turtles. Today, ebuilds don't even let a chance for an admin to apply a series of patches to the vanilla/distro-maintainer sources without having to rewrite/fork the ebuild. There isn't a way to specify ebuild properties in a way like command line arguments? Where you can explicitly silence options by specifying them later, etc? On Sun, Nov 23, 2014 at 6:30 PM, Rich Freeman ri...@gentoo.org wrote: On Sun, Nov 23, 2014 at 7:12 PM, hasufell hasuf...@gentoo.org wrote: On 11/24/2014 12:24 AM, Rich Freeman wrote: * kickban major assholes from the community, no matter how efficient they are Proposals welcome. Hint, things will go much better if you volunteer to do the work the assholes are doing... It isn't like we aren't all tired of this stuff, but if we go booting half the devs then the distro will basically die. That's actually an argument FOR my proposal of being more distributed and shrinking the dev community. In such a scenario we would not need 200 gentoo developers anymore. Sure, but my point is that the way to fix this is: 1. Set up new distributed model. 2. Work in the new model successfully for a while. 3. Retire developers who are no longer needed since they won't be doing anything anyway. And not: 1. Stop recruiting new devs. 2. Watch attrition get rid of existing devs. 3. Work on new distributed model that may or may not ever take off. 4. Hope that Gentoo doesn't die in the meantime. -- Rich
Re: [gentoo-user] The future of linux, and Gentoo specifically now
The reason this question is so hard to answer is because it is not a technical question, it is a moral and ethical one. The links presented start to approach the issue being discussed in this light but do not entirely accept the right question. I suspect this is because it seems rather absurd. We shall analyze some popular responses in this light. Systemd is easy to work around! http://www.vitavonni.de/blog/201410/2014102101-avoiding-systemd.html except, https://lobste.rs/s/y5skqt/avoiding_systemd_isn_t_hard/comments/eayjn3#c_eayjn3 but http://0pointer.de/blog/projects/the-biggest-myths.html gives some decent counterpoints, which http://lwn.net/Articles/619992/ either supports or is ambivalent about. They all basically boil down to someone is doing the work, and if it is a better way to do it it will be okay. Except this isn't true. The proof by contradiction is exceptionally simple: If this was a just world, Lennart's pants would be on fire. Lennart's pants are not on fire. Therefore, this is not a just world, and justice must be manufactured. You might ask why his pants (and the pants of most systemd supporters) would be on fire. Well, https://pappp.net/?p=969 clearly explains how FLOS is not UNIX, and the easy counterpoints get thoroughly trashed http://lwn.net/Articles/440843/, and http://blog.lusis.org/blog/2014/11/20/systemd-redux/ here's a guy agreeing and suggesting everyone hit the big red EJECT. Why UNIX? Well, because that's just a concise, easy-to-phrase proxy for the deeper issue of https://pay.reddit.com/r/linux/comments/2k5b7e/the_concern_isnt_that_systemd_itself_isnt/ (aside: read the C++ in the kernel tangent if you are not familiar, it seems to mirror this argument taking place and notably, Linus has chosen a side on that one!) which is echoed here http://lwn.net/Articles/440843/ and here http://lwn.net/Articles/576078/ and here http://uselessd.darknedgy.net/ProSystemdAntiSystemd/ (start with unix philosophy) and here http://lwn.net/Articles/494605/. Once upon a time I met a very masterful troll who got me to say precisely what I needed to say precisely when I did not want to say it. What he got me to say was: Oct 27 06:05:30 ***I study the orthodoxy consistently[sic] Oct 27 06:05:38 R0b0t1`To find its flaws, yes So did Lennart co. study the orthodox to learn from its failures? Did they construct a conservative (re)implementation of the software exhibiting those failures? It has been shown and continues to be shown that: no, they are flying by the seat of their pants. A solution could have been constructed which requires far less labor. Not only far less of *their* labor, but far less labor for *everyone else* using a *nix. But they did not thoroughly investigate such avenues, even within their reimplementation! They are recreating bugs! It is impossible for them to claim they are doing it over to do it right, as they have already failed at that purpose. They have been shown to have wasted effort and continue to do so. When labor is scarce, that is the most unethical action one can undertake. On Fri, Nov 21, 2014 at 4:04 PM, Marc Joliet mar...@gmx.de wrote: Am Fri, 21 Nov 2014 01:32:16 -0600 schrieb Canek Peláez Valdés can...@gmail.com: [...] I highly recommend the article John Corbet wrote for LWN a week ago: http://lwn.net/Articles/619992/ [...] Thanks for the link, it was a good read. FWIW, I found this linked in one of the comments: http://uselessd.darknedgy.net/ProSystemdAntiSystemd/ Both articles echo thoughts that I have more and more with every discussion regarding systemd. My takeaway is similar to that of the lwn.net article (that is, both sides are being unnecessarily thick-headed), and find it remarkable how much I recognise from discussions here on gentoo-user (in contrast, gentoo-amd64 has been much more level-headed). However, I disagree with with the categorisation at the end, mainly because I hate it when people have to sort each other into camps, so that they know who to hate and who to like (which isn't the author's fault, I think, politicised discussions tend to go that way as they intensify), but also because I think it is too strict and doesn't account for overlap (for myself I see reasons for both being and not being in either group). Greetings -- Marc Joliet -- People who think they know everything really annoy those of us who know we don't - Bjarne Stroustrup
Re: [gentoo-user] The future of linux, and Gentoo specifically now
Regardless, it would probably be useful to contact the people from the Debian project who were interested in forking it. It's likely Gentoo would end up using a fair amount of their work at some point. On Mon, Nov 24, 2014 at 6:18 AM, Sid S r03...@gmail.com wrote: The reason this question is so hard to answer is because it is not a technical question, it is a moral and ethical one. The links presented start to approach the issue being discussed in this light but do not entirely accept the right question. I suspect this is because it seems rather absurd. We shall analyze some popular responses in this light. Systemd is easy to work around! http://www.vitavonni.de/blog/201410/2014102101-avoiding-systemd.html except, https://lobste.rs/s/y5skqt/avoiding_systemd_isn_t_hard/comments/eayjn3#c_eayjn3 but http://0pointer.de/blog/projects/the-biggest-myths.html gives some decent counterpoints, which http://lwn.net/Articles/619992/ either supports or is ambivalent about. They all basically boil down to someone is doing the work, and if it is a better way to do it it will be okay. Except this isn't true. The proof by contradiction is exceptionally simple: If this was a just world, Lennart's pants would be on fire. Lennart's pants are not on fire. Therefore, this is not a just world, and justice must be manufactured. You might ask why his pants (and the pants of most systemd supporters) would be on fire. Well, https://pappp.net/?p=969 clearly explains how FLOS is not UNIX, and the easy counterpoints get thoroughly trashed http://lwn.net/Articles/440843/, and http://blog.lusis.org/blog/2014/11/20/systemd-redux/ here's a guy agreeing and suggesting everyone hit the big red EJECT. Why UNIX? Well, because that's just a concise, easy-to-phrase proxy for the deeper issue of https://pay.reddit.com/r/linux/comments/2k5b7e/the_concern_isnt_that_systemd_itself_isnt/ (aside: read the C++ in the kernel tangent if you are not familiar, it seems to mirror this argument taking place and notably, Linus has chosen a side on that one!) which is echoed here http://lwn.net/Articles/440843/ and here http://lwn.net/Articles/576078/ and here http://uselessd.darknedgy.net/ProSystemdAntiSystemd/ (start with unix philosophy) and here http://lwn.net/Articles/494605/. Once upon a time I met a very masterful troll who got me to say precisely what I needed to say precisely when I did not want to say it. What he got me to say was: Oct 27 06:05:30 ***I study the orthodoxy consistently[sic] Oct 27 06:05:38 R0b0t1`To find its flaws, yes So did Lennart co. study the orthodox to learn from its failures? Did they construct a conservative (re)implementation of the software exhibiting those failures? It has been shown and continues to be shown that: no, they are flying by the seat of their pants. A solution could have been constructed which requires far less labor. Not only far less of *their* labor, but far less labor for *everyone else* using a *nix. But they did not thoroughly investigate such avenues, even within their reimplementation! They are recreating bugs! It is impossible for them to claim they are doing it over to do it right, as they have already failed at that purpose. They have been shown to have wasted effort and continue to do so. When labor is scarce, that is the most unethical action one can undertake. On Fri, Nov 21, 2014 at 4:04 PM, Marc Joliet mar...@gmx.de wrote: Am Fri, 21 Nov 2014 01:32:16 -0600 schrieb Canek Peláez Valdés can...@gmail.com: [...] I highly recommend the article John Corbet wrote for LWN a week ago: http://lwn.net/Articles/619992/ [...] Thanks for the link, it was a good read. FWIW, I found this linked in one of the comments: http://uselessd.darknedgy.net/ProSystemdAntiSystemd/ Both articles echo thoughts that I have more and more with every discussion regarding systemd. My takeaway is similar to that of the lwn.net article (that is, both sides are being unnecessarily thick-headed), and find it remarkable how much I recognise from discussions here on gentoo-user (in contrast, gentoo-amd64 has been much more level-headed). However, I disagree with with the categorisation at the end, mainly because I hate it when people have to sort each other into camps, so that they know who to hate and who to like (which isn't the author's fault, I think, politicised discussions tend to go that way as they intensify), but also because I think it is too strict and doesn't account for overlap (for myself I see reasons for both being and not being in either group). Greetings -- Marc Joliet -- People who think they know everything really annoy those of us who know we don't - Bjarne Stroustrup
Re: [gentoo-user] Gentoo's future directtion ?
Oh. I've had to use that, even. I was thinking patches of ebuilds. (???) On Mon, Nov 24, 2014 at 6:41 AM, Rich Freeman ri...@gentoo.org wrote: On Mon, Nov 24, 2014 at 3:20 AM, Sid S r03...@gmail.com wrote: Today, ebuilds don't even let a chance for an admin to apply a series of patches to the vanilla/distro-maintainer sources without having to rewrite/fork the ebuild. There isn't a way to specify ebuild properties in a way like command line arguments? Where you can explicitly silence options by specifying them later, etc? Any environment-level property can be overridden at the command line, though this does not include patching. However, the original claim is still wrong - you just stick the patches in /etc/portage/patches. This only works for ebuilds that call epatch_user right now, but for EAPI6 it will work for all packages. -- Rich
Re: [gentoo-user] OT: new thinkpad with Gentoo
I will agree with the suggestion that Asus laptop keyboards are decent. On the higher end models the keys have a surprising amount of travel. As for keyboards with a trackpoint, I would suggest the TEX Yoda Trackpoint and the Miniguru keyboards. Sadly, neither is available with any regularity (the second seems to have had a prototype run but is currently in design), and I believe the mouse is different from a traditional trackpoint by virtue of patents. I became very accustomed to the trackpoint on a previous laptop I had. Since they are so rare, I've turned to tiling window managers to remove my need of mouse. On Fri, Nov 21, 2014 at 2:45 PM, Stefan G. Weichinger li...@xunil.at wrote: Am 20.11.2014 um 21:16 schrieb Daniel Frey: On 11/20/2014 11:16 AM, thegeezer wrote: yeah at first it's odd, but then when you start getting used to navigating without removing hands from keyboard it does become almost a prerequisite. does anyone know if you can get usb keyboards that have the trackpoint style mini-joystick in the middle of them ? Yep: http://shop.lenovo.com/us/en/itemdetails/0B47190/460/60AC6A0372B14F5BA7B12F1FF88E33C7 I almost bought this one but I wanted a usb port on my keyboard itself for my mouse, IIRC this one didn't have that. I have used a lenovo keyboard with it, I liked it, just wish it had a USB port for the mouse. I am using a Lenovo keyboard (USB Keyboard SK-8815) for my main workstation for years now. Still working fine. It does not bring a trackpoint but has 2 USB ports and some special function keys I never managed to get working ... maybe I should take another approach as it might be supported for months and years now. Stefan
Re: [gentoo-user] OT: new thinkpad with Gentoo
In a similar vein, I would suggest https://system76.com/laptops. I found them after I purchased my laptop. Had I known, I likely would have gone with them and purchased their most expensive model. For the most part, what you want is relatively hard to get with a typical consumer computer. They've gone for minimization to the point where they can't offer (or don't want to offer) the options that the majority of people would expect to exist. On Wed, Nov 19, 2014 at 3:08 AM, J. Roeleveld jo...@antarean.org wrote: On Tuesday, November 18, 2014 11:45:05 PM Stefan G. Weichinger wrote: I consider buying a new laptop in late 2014 ... taxes and stuff ... I run 2 thinkpads here, each with 8 gigs of RAM and SSD inside: L520 and X220 - both still with Intel Core i-(5|7)-2xxx inside. So far OK, but not up2date. Considering a budget of ~1000 EUR maximum (more based on reason) ... do you have any recommendation from practical experience? I once had a Macbook Air (1st generation) with gentoo on it, it was nice because it was slim and lightweight ... a similar device costs 2000 bucks up when I buy it new (Lenovo Thinkpads X... or X1 Carbon). What I want to avoid is spending quite some money and getting basically the same weight/size I already know with minor performance gain (which I get used to within days anyway). opinions appreciated! :-) Stefan Stefan, Instead of going with the big brands, you could also go for one of the lesser known brands. I myself am using a laptop from BTO (www.bto.eu). They actually are based near my home, which saved me the shipping costs. With this company (and similar ones like it), you have a lot of freedom when deciding on what ends up inside it. A lot more then when looking for laptops from the big brands. And, they're also cheaper. -- Joost
Re: [gentoo-user] Anybody using a yubikey neo as Smartcard as well?
No, but I was interested in getting one. Did you configure it properly? https://www.yubico.com/2012/12/yubikey-neo-composite-device/ https://developers.yubico.com/libykneomgr/ On Sat, Nov 15, 2014 at 5:24 PM, Konstantinos Agouros elw...@agouros.de wrote: Hi, I just got a yubikey neo that I would like to use as a smart card as well as just for OTP. However the pscsd complains, that it can not talk to it, although it tries to. The error I get starts with: Nov 16 00:23:17 rumba pcscd: /var/tmp/portage/sys-apps/pcsc-lite-1.8.13-r1/work/pcsc-lite-1.8.13/src/readerfactory.c:1043:RFInitializeReader() Open Port 0x20 Failed (usb:1050/0111:libhal:/org/freedesktop/Hal/devices/usb_device_1050_0111_serialnotneeded_if0) I tried compiling with libusb as well as with udev USE flags the error stays the same. Any recommendations? Konstantin -- Dipl-Inf. Konstantin Agouros aka Elwood Blues. Internet: elw...@agouros.de Altersheimerstr. 1, 81545 Muenchen, Germany. Tel +49 89 69370185 There is no 'dead' in team! - Sameen Shaw
Re: [gentoo-user] Running a program on a headless computer ?
I was deciding whether to reply earlier, and my worst fears were realized: you plan to start this automatically with no human intervention. Here is a snippet which should do what you want: `nohup my-program /dev/null 21 ` Screen and tmux will/should work, but at the very least I would suggest wrapping the program with a program of your own where you execute the other and have more and explicit control over the file descriptors. The best solution is modifying your programs to work headlessly. This also means you can ensure logging is done properly (you were logging, right?), and not in some ad-hoc fashion like reading all the messages your black box produces and keeping the ones that look important. It's tempting to glue some things together with the shell, but I would avoid reliance on programs which are not yours as much as possible. Not Invented Here is often (rightfully) mocked, but this is one of the situations in which it has some relevance. On Sun, Sep 28, 2014 at 2:15 PM, Todd Goodman t...@bonedaddy.net wrote: * meino.cra...@gmx.de meino.cra...@gmx.de [140928 10:14]: Hi, I want to run programs, which insist on haveing a terminal to write their status to and which are writing files which their results on a headless computer (beaglebone). I tried things like my_program -o file.txt -parameter value /dev/null 21 but this results in a idle copy of this process and a defunct child. The program does not use X11 in any way... Is there any neat trick to accomplish what I am trying to do here? Thank you very much in advance for any help! Best regards, mcc You probably want 21
Re: [gentoo-user] installed Gentoo on SSD - no bootable device
I believe what you've said is correct... because I'm pretty sure I read it in the documentation. On Thu, Sep 4, 2014 at 11:08 PM, Joseph syscon...@gmail.com wrote: On 09/04/14 20:44, Daniel Frey wrote: On 09/04/2014 08:14 PM, Joseph wrote: On 09/04/14 19:41, Daniel Frey wrote: On 09/04/2014 05:36 PM, Joseph wrote: When I installed grub2 I got no errors: grub2-install /dev/sda Installation finished. No error reported. If you are trying to boot in EFI mode, you aren't installing it correctly. That installed to the MBR in legacy mode. Instructions are here: http://wiki.gentoo.org/wiki/GRUB2 You need to mount /boot, and mount the EFI boot partition before installing grub2 using `grub2-install --target=x86_64-efi`. I'm still lost with this grab2, very confusing. Gentoo official documentation did not mention any of this :-/ Official documentation did ask to create /dev/sda1 2M BIOS boot partition but there was no instruction how to mount it or format it. I was under impression Grub2 will do all of this. I booted with CD-minimal and there is no mkdosfs command. Do I need to format the /dev/sda1? If I do: mkfs -t vfat -F 32 -n efi-boot /dev/sda1 mkfs.vfat: No such file or directory I forgot to mention in my last post that you absolutely must boot from an EFI-enabled kernel, the gentoo ISOs do not do this. I used the Mint 17 ISO to do this, when you go to boot options it should list it as EFI bootable. Thank you for explanation. I have a question with regards to that EFI. Does it refer to this /dev/sda1 2M BIOS boot partition? So this partition needs to be formatted to DOS file system and mounted in /boot/efi directory? Gentoo Documentation is very outdated and confusing when it comes to this new GRUB2. Sometimes I want to scrap this crap and go back to standard legacy GRUB. -- Joseph
Re: [gentoo-user] SSD recmmendation / input
An SSD has the best performance return per dollar than most any other investment you can make (for a typical workload). It's actually rather unlikely you'll get a bad one if you stick to relatively known brands. ADATA has some cheap options that, while pretty low on the totem poll, will give you a decent amount of storage and speed. If you don't mind spending more you can look at Crucial, PNY, Intel, etc. But keep in mind you'll receive diminishing returns w.r.t. higher read and write speeds - you actually get most of the speed increase due to an SSD's better random read response, which can be had with even the cheapest ones. Any disk you pick will be compatible with your motherboard, though if it is older you may not reach the SSD's peak throughput. I use a 256gb SSD drive with two 1TB platter drives. It will hold your OS, swap, and some media. I use LVM. Wear levelling will work properly even if you partition the whole disk On Fri, Aug 29, 2014 at 1:14 PM, cov...@ccs.covici.com wrote: Joseph syscon...@gmail.com wrote: The Crucial 512GB SSD is not that expensive and I found some notes on partitioning SSD on Gentoo: https://wiki.gentoo.org/wiki/SSD It seems to me I'll only have boot, swap and root partition; home I think will be mounted on root partition. -- Joseph On 08/29/14 07:49, Daniel Frey wrote: On 08/28/2014 09:54 PM, Joseph wrote: No, I wouldn't get 1TB SSD too expensive but something like 300GB I might consider it. Are they worth the investment? What brand do you have and how long? I have several SSDs. I currently use Kingston, Crucial, and Intel. A bit of background - I use a mythtv setup with multiple frontends. I had a SSD in the backend but it failed after about two years with no warning -- one day I noticed the frontends behaving strangely and found out I couldn't log into the backend (via ssh or directly.) The server sustained a lot of writes to the database daily, however, the actual recordings were on rust disks. It was a Kingston that failed, a 32GB model. The Crucial and Intel I have are still relatively new, the Crucial being a year and a bit old, and the Intel only a few days old. :-) Speedwise, there's no comparison. Especially running emerge/compiling - my frontend (equipped with an E8400 and 2GB RAM) with the Kingston SSD beats my main workstation equipped with a rust raid10 (a QX9650 with 8 GB RAM) every time. I have two recommendations for a new SSD user - 1) Flash the firmware to a new version right away if available, and 2) Don't partition the entire SSD if you can avoid it. Apparently SSDs will use unused space for wear leveling - as an example I believe I only partitioned 20GB (out of a 64GB SSD) on my frontends. That's a bit excessive and you may not be able to do that, but you get the idea. Also make sure to use parted to partition so the partitions themselves are aligned properly. (Regarding the firmware update - my Crucial had one and I ignored it. About 3 months later my laptop was acting weird and complaining about the disk. I was lucky - I flashed the firmware and it was fine with no data loss. Others are not so lucky...) How is the partitioning advise effected by lvm? I use that all the time and just do a normal boot partition and the rest given over to lvm. But this may be not good with an ssd. -- Your life is like a penny. You're going to lose it. The question is: How do you spend it? John Covici cov...@ccs.covici.com
Re: [gentoo-user] glibc (and gcc) build fails: /bin/sh: /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/cross-rpcgen: No such file or directory
Message delivery has been failing, sorry if this is received twice. Wow, running Steam on Hardened. Seems ambitious. Most things work after grsec is clubbed over the head (sadly the only real option in a lot of cases). 1. Reply with a list of actions/commands you did that led up to this point I untared a file extracted from the Valve-provided .deb to /. I then created the package set and emerged it; it failed while compiling glibc. Something during these steps apparently broke glibc/gcc (with a different error than I am asking about now), so I had to reinstall them from http://tinderbox.dev.gentoo.org and copy a bunch of headers from the 64 bit directory to a 32 bit one. Now I'm getting this error. I think part of it goes back to when I initially extracted the Valve-provided .tar.gz... It seems to have overwritten a lot of files, but I assumed and have been told this is not the default behavior, or even really possible (it would have had to clear pre-existing directories). Do you mean I tried to install @steam and it failed once it got to glibc, or were you running with --keep-going and figured it out in hindsight? It failed immediately, I did not use --keep-going. Did you do a deep update before doing all of this? Yes. 2. Attach the full build log It's too long to include in the message body. What is the best way to send it? If you do not mind visiting an external link, here is one: http://bpaste.net/show/488606/. On Tue, Jul 22, 2014 at 11:26 AM, Alec Ten Harmsel a...@alectenharmsel.com wrote: Wow, running Steam on Hardened. Seems ambitious. I'll try and help as much as possible. Can you: 1. Reply with a list of actions/commands you did that led up to this point 2. Attach the full build log Also They originally got rebuilt (and I noticed the failure) Do you mean I tried to install @steam and it failed once it got to glibc, or were you running with --keep-going and figured it out in hindsight? Did you do a deep update before doing all of this? I've only been running Gentoo for a couple of years, so sorry for asking all these questions. Most of the time a build fails on my machine it's because I've done something really, really stupid, so just don't want to make the mistake of diving in way too deep and then figuring out it was a simple issue. Alec On Tue 22 Jul 2014 12:15:08 PM EDT, Sid S wrote: Are you in the process of switching to hardened right now? Why are you rebuilding glibc and gcc? No, I set up my system as hardened. They originally got rebuilt (and I noticed the failure) when I set up a package set for running some games (using the steam client) and rebuilt the package set. They are the packages listed here: http://wiki.gentoo.org/wiki/Steam. The file you sent would have been half the size, if you'll excuse me nagging, if you'd sent it in plaintext, rather than HTML. Did something strange happen? I figured the code blocks would just be surrounded by styling tags. Sorry about that. On Tue, Jul 22, 2014 at 10:36 AM, Stroller strol...@stellar.eclipse.co.uk mailto:strol...@stellar.eclipse.co.uk wrote: On Tue, 22 July 2014, at 2:48 pm, Sid S r03...@gmail.com mailto:r03...@gmail.com wrote: ... The build log is kind of large, tell me if the whole thing is needed. The file you sent would have been half the size, if you'll excuse me nagging, if you'd sent it in plaintext, rather than HTML. Stroller.
[gentoo-user] glibc (and gcc) build fails: /bin/sh: /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/cross-rpcgen: No such file or directory
Not sure what relevant information to provide. My system works and I can compile other ebuilds normally, except, it seems, gcc and glibc. Searching for the error in the tile doesn't give me much - what it does give me doesn't seem to apply to my instance of the error, or at least what I know about it (patches, etc). A common solution is to switch to a non-multilib profile, but I need multilib. I am using the hardened profile with selinux, but it doesn't seem to be an issue with either of those (it's usually pretty obvious if it is). `MAKEOPTS=-j1 emerge glibc`: x86_64-pc-linux-gnu-gcc -m32 rpcgen.c -c -std=gnu99 -fgnu89-inline -O2 -Wall -Winline -Wwrite-strings -fmerge-all-constants -fno-stack-protector -fno-strict-aliasing -frounding-math -pipe -Wstrict-prototypes -Wa,-mtune=i686-U_FORTIFY_SOURCE -DPIC -DPIC -DPIC -D_RPC_THREAD_SAFE_ -I../include -I/var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc -I/var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl -I../nptl/sysdeps/unix/sysv/linux/i386/i686 -I../sysdeps/unix/sysv/linux/i386/i686 -I../nptl/sysdeps/unix/sysv/linux/i386 -I../nptl/sysdeps/unix/sysv/linux/x86 -I../sysdeps/unix/sysv/linux/x86 -I../sysdeps/unix/sysv/linux/i386/nptl -I../sysdeps/unix/sysv/linux/i386 -I../nptl/sysdeps/unix/sysv/linux -I../nptl/sysdeps/pthread -I../sysdeps/pthread -I../ports/sysdeps/unix/sysv/linux -I../sysdeps/unix/sysv/linux -I../sysdeps/gnu -I../sysdeps/unix/inet -I../nptl/sysdeps/unix/sysv -I../ports/sysdeps/unix/sysv -I../sysdeps/unix/sysv -I../sysdeps/unix/i386 -I../nptl/sysdeps/unix -I../ports/sysdeps/unix -I../sysdeps/unix -I../sysdeps/posix -I../sysdeps/i386/i686/fpu/multiarch -I../sysdeps/i386/i686/fpu -I../sysdeps/i386/i686/multiarch -I../nptl/sysdeps/i386/i686 -I../sysdeps/i386/i686 -I../sysdeps/i386/i486 -I../nptl/sysdeps/i386/i486 -I../sysdeps/i386/fpu -I../sysdeps/x86/fpu -I../nptl/sysdeps/i386 -I../sysdeps/i386 -I../sysdeps/x86 -I../sysdeps/wordsize-32 -I../sysdeps/ieee754/ldbl-96 -I../sysdeps/ieee754/dbl-64 -I../sysdeps/ieee754/flt-32 -I../sysdeps/ieee754 -I../sysdeps/generic -I../nptl -I../ports -I.. -I../libio -I. -nostdinc -isystem /usr/lib/gcc/x86_64-pc-linux-gnu/4.7.3/../../../../lib32/include -isystem /usr/lib/gcc/x86_64-pc-linux-gnu/4.7.3/include-fixed -isystem /usr/include -D_LIBC_REENTRANT -include ../include/libc-symbols.h -DNOT_IN_libc=1-D_RPC_THREAD_SAFE_ -o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpcgen.o -MD -MP -MF /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpcgen.o.dt -MT /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpcgen.o x86_64-pc-linux-gnu-gcc -m32 -pie -Wl,-O1 -nostdlib -nostartfiles -o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpcgen -Wl,-dynamic-linker=/lib32/ld-linux.so.2 -Wl,-O1 -Wl,--as-needed -Wl,-z,combreloc -Wl,-z,relro /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/csu/Scrt1.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/csu/crti.o `x86_64-pc-linux-gnu-gcc -m32 --print-file-name=crtbeginS.o` /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpcgen.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_main.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_hout.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_cout.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_parse.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_scan.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_util.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_svcout.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_clntout.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_tblout.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_sample.o
Re: [gentoo-user] glibc (and gcc) build fails: /bin/sh: /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/cross-rpcgen: No such file or directory
I have IA32 emulation enabled but none of its suboptions (I read I don't need them). Sorry about the lines running off, my client does not auto-wrap with newlines and it seems yours does not either. . On Tue, Jul 22, 2014 at 8:58 AM, Alec Ten Harmsel a...@alectenharmsel.com wrote: I've had some problems with this in the past on non-hardened... you have IA32 emulation turned on in the kernel right? Also, not sure if this is just me, but the really long lines (like the top 4) run off the end of my screen and I can't see them at all. Alec On Tue 22 Jul 2014 09:48:54 AM EDT, Sid S wrote: Not sure what relevant information to provide. My system works and I can compile other ebuilds normally, except, it seems, gcc and glibc. Searching for the error in the tile doesn't give me much - what it does give me doesn't seem to apply to my instance of the error, or at least what I know about it (patches, etc). A common solution is to switch to a non-multilib profile, but I need multilib. I am using the hardened profile with selinux, but it doesn't seem to be an issue with either of those (it's usually pretty obvious if it is). `MAKEOPTS=-j1 emerge glibc`: x86_64-pc-linux-gnu-gcc -m32 rpcgen.c -c -std=gnu99 -fgnu89-inline -O2 -Wall -Winline -Wwrite-strings -fmerge-all-constants -fno-stack-protector -fno-strict-aliasing -frounding-math -pipe -Wstrict-prototypes -Wa,-mtune=i686-U_FORTIFY_SOURCE -DPIC -DPIC -DPIC -D_RPC_THREAD_SAFE_ -I../include -I/var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc -I/var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl -I../nptl/sysdeps/unix/sysv/linux/i386/i686 -I../sysdeps/unix/sysv/linux/i386/i686 -I../nptl/sysdeps/unix/sysv/linux/i386 -I../nptl/sysdeps/unix/sysv/linux/x86 -I../sysdeps/unix/sysv/linux/x86 -I../sysdeps/unix/sysv/linux/i386/nptl -I../sysdeps/unix/sysv/linux/i386 -I../nptl/sysdeps/unix/sysv/linux -I../nptl/sysdeps/pthread -I../sysdeps/pthread -I../ports/sysdeps/unix/sysv/linux -I../sysdeps/unix/sysv/linux -I../sysdeps/gnu -I../sysdeps/unix/inet -I../nptl/sysdeps/unix/sysv -I../ports/sysdeps/unix/sysv -I../sysdeps/unix/sysv - I ../sysdeps/unix/i386 -I../nptl/sysdeps/unix -I../ports/sysdeps/unix -I../sysdeps/unix -I../sysdeps/posix -I../sysdeps/i386/i686/fpu/multiarch -I../sysdeps/i386/i686/fpu -I../sysdeps/i386/i686/multiarch -I../nptl/sysdeps/i386/i686 -I../sysdeps/i386/i686 -I../sysdeps/i386/i486 -I../nptl/sysdeps/i386/i486 -I../sysdeps/i386/fpu -I../sysdeps/x86/fpu -I../nptl/sysdeps/i386 -I../sysdeps/i386 -I../sysdeps/x86 -I../sysdeps/wordsize-32 -I../sysdeps/ieee754/ldbl-96 -I../sysdeps/ieee754/dbl-64 -I../sysdeps/ieee754/flt-32 -I../sysdeps/ieee754 -I../sysdeps/generic -I../nptl -I../ports -I.. -I../libio -I. -nostdinc -isystem /usr/lib/gcc/x86_64-pc-linux-gnu/4.7.3/../../../../lib32/include -isystem /usr/lib/gcc/x86_64-pc-linux-gnu/4.7.3/include-fixed -isystem /usr/include -D_LIBC_REENTRANT -include ../include/libc-symbols.h -DNOT_IN_libc=1 -D_RPC_THREAD_SAFE_ -o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpcgen.o -MD -MP -MF /var/tmp/portage/sys-libs/gl i bc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpcgen.o.dt -MT /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpcgen.o x86_64-pc-linux-gnu-gcc -m32 -pie -Wl,-O1 -nostdlib -nostartfiles -o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpcgen -Wl,-dynamic-linker=/lib32/ld-linux.so.2 -Wl,-O1 -Wl,--as-needed -Wl,-z,combreloc -Wl,-z,relro /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/csu/Scrt1.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/csu/crti.o `x86_64-pc-linux-gnu-gcc -m32 --print-file-name=crtbeginS.o` /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpcgen.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_main.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_hout.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_cout.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_parse.o /var/tmp/portage/sys-libs / glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_scan.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_util.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_svcout.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_clntout.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_tblout.o /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/rpc_sample.o -Wl,-rpath-link=/var/tmp
Re: [gentoo-user] glibc (and gcc) build fails: /bin/sh: /var/tmp/portage/sys-libs/glibc-2.17/work/build-x86-x86_64-pc-linux-gnu-nptl/sunrpc/cross-rpcgen: No such file or directory
Are you in the process of switching to hardened right now? Why are you rebuilding glibc and gcc? No, I set up my system as hardened. They originally got rebuilt (and I noticed the failure) when I set up a package set for running some games (using the steam client) and rebuilt the package set. They are the packages listed here: http://wiki.gentoo.org/wiki/Steam. The file you sent would have been half the size, if you'll excuse me nagging, if you'd sent it in plaintext, rather than HTML. Did something strange happen? I figured the code blocks would just be surrounded by styling tags. Sorry about that. On Tue, Jul 22, 2014 at 10:36 AM, Stroller strol...@stellar.eclipse.co.uk wrote: On Tue, 22 July 2014, at 2:48 pm, Sid S r03...@gmail.com wrote: ... The build log is kind of large, tell me if the whole thing is needed. The file you sent would have been half the size, if you'll excuse me nagging, if you'd sent it in plaintext, rather than HTML. Stroller.
