- Arch-general - lists.archlinux.org

[arch-general] clients can no longer mount.cifs the '/' samba share from current Arch server (long)
by David C. Rankin 12 Jan '14

12 Jan '14

Guys, Something in current Archlinux prevents clients mounting a root samba share '/' via mount.cifs. Since building a new Arch server a month or so ago, I have been chasing an issue with mount.cifs that prevents the '/' share from being mounted as it always has. This has worked (and continues to work) on all Arch boxes up until this latest box. I have addressed this to the samba list (see thread: [Samba] What in samba 4.1 prevents a '/' share?) After a month of troubleshooting with the samba devs, it seems that the issue may have to do with the way /proc/fs/cifs/SecurityFlags, or some other default is now set in Arch. Others have confirmed the ability to mount '/' shares from servers running samba 4.1.3, with the exact smb.conf settings, so it appears this is Arch specific. I'm not 100% sure if this is samba related or whether it is cifs related, so I am asking here so I can get the bug report right -- if needed. Several bug reports deal with the drop of ntlm security for the kernel as of kernel 3.8 and with differing /proc/fs/cifs/SecurityFlags values -- https://bugs.launchpad.net/ubuntu/+source/cifs-utils/+bug/1113395 (see specifically No. #12 & #15 citing: https://bbs.archlinux.org/viewtopic.php?id=159915) Historically the value of /proc/fs/cifs/SecurityFlags has been set at 0x7 (or not set at all on Arch), but now Arch sets the value to 0x85. Examples: (older Arch box smbd Version 3.6.6) [19:32 nirvana:/etc] # cat /proc/fs/cifs/SecurityFlags cat: /proc/fs/cifs/SecurityFlags: No such file or directory (new Arch box smbd Version 4.1.3) [19:30 phoinix:/home/david/cnf/phoinix/etc] # cat /proc/fs/cifs/SecurityFlags 0x85 I have set /proc/fs/cifs/SecurityFlags to the traditional value of 0x7, but there is still something that prevents the mount from working. I will detail below the config and attempts made to resolve the issue. The bottom line is that all older Arch servers (and openSuSE boxes) I have can successfully share/mount a '/' root samba share, but there is now something in current Arch that prevents this from working. The folks at the samba list are stumped because there is nothing in samba or cifs that has changed that would prevent this from working. The bug reports I've read suggest the cifs/SecurityFlags setting may be involved, but from the testing I've done, I don't know if that makes any difference. Now I need help from you smarter Archers to help figure out what changed that prevents this from working and how to fix it. The current Arch box (hostname phoinix) is a fully updated box (as of 1/6/14) with: # smbd --version Version 3.6.7 # mount.cifs -V mount.cifs version: 5.6 The samba config: The server functions as a standalone server (no PDC, etc..). The smb.conf and share definitions are: Server role: ROLE_STANDALONE Press enter to see a dump of your service definitions [global] workgroup = RLFPLLC server string = Phoinix Samba %v map to guest = Bad User smb passwd file = /etc/samba/smbpasswd log file = /var/log/samba/%m.log max log size = 50 time server = Yes printcap name = /etc/printcap show add printer wizard = No os level = 66 preferred master = Yes domain master = Yes dns proxy = No wins support = Yes idmap config * : backend = tdb admin users = david hosts allow = 192.168.6., 192.168.7., 127. use sendfile = Yes [homes] comment = Home Directories read only = No browseable = No [config] comment = Phoinix Config (Archlinux) path = / valid users = david force user = root force group = root read only = No browseable = No [samba] comment = Phoinix - Law path = /home/samba valid users = @rankin, #, anna force group = rankin read only = No inherit permissions = Yes <snip> Mounting the homes and samba shares work fine and printing via cups works fine. The only issue is mounting the [config] share. The shares are mounted making use of a credentials file with: mount.cifs //phoinix/config /mnt/phx-cfg -v -o username=david,uid=1000,credentials=/home/david/.dcr/mountcfile,noperm The results of mount showing successful mount of the [homes] and [samba] shares: //phoinix/samba on /mnt/phx type cifs (rw,relatime,vers=1.0,sec=ntlm,cache=loose,unc=\\phoinix\samba,username=david,uid=1000,forceuid,gid=0,noforcegid,addr=192.168.7.16,unix,posixpaths,serverino,acl,noperm,rsize=1048576,wsize=65536,actimeo=1) //phoinix/david on /mnt/phx-david type cifs (rw,relatime,vers=1.0,sec=ntlm,cache=loose,unc=\\phoinix\david,username=david,uid=1000,forceuid,gid=0,noforcegid,addr=192.168.7.16,unix,posixpaths,serverino,acl,noperm,rsize=1048576,wsize=65536,actimeo=1) However, attempting to mount the [config] share results in the error: [18:33 providence:~/tmp/cifs] # mount.cifs //phoinix/config /mnt/phx-cfg -v -o username=david,uid=1000,credentials=/home/david/.dcr/mountcfile,noperm mount.cifs kernel mount options: ip=192.168.7.16,unc=\\phoinix\config,noperm,uid=1000,user=david,pass=******** mount error(13): Permission denied Refer to the mount.cifs(8) manual page (e.g. man mount.cifs) I have tried with 'sec=ntlmv2' and 'sec=ntlmssp' which makes no difference. I have tried with/without 'domain=rlfpllc' again no difference. What I need to determine is whether this is a bug or whether this is a configuration issue, and if so, in what? I have captured tcpdump traffic during the mount attempts and they point to smb issuing the error, but I'm not that great at reading packet contents, so I'm not entirely sure. But basically, after successful AndX session setup (Tree Connect AndX Request, Path: \\phoinix\config), the request for \\phoinix\config is made and it is found successfully by the server, but then the server response with (Tree Connect AndX Response, Error: STATUS_ACCESS_DENIED) The full ASCII dump of the packet along with additional testing showing it works on all older servers is included below. If you have ideas or would like me to post additional information, just let me know. I have worked to collect the relevant information from the samba thread, but let me know if you need anything else Arch related from the box. Thanks in advance for any help you can provide. ====== Additional Testing and ASCII dump of STATUS_ACCESS_DENIED packet ====== Testing with other servers I can easily mount the '/' share from any other computer. Here is my 3-computer test setup: On my client (hostname providence) I have: 18:01 providence:~> smbd -V Version 3.6.7 linux 3.5.3-1 cifs-utils 5.6-2 On the server (hostname phoinix) I have: 18:15 phoinix:~> smbd -V Version 4.1.3 linux 3.12.6-1 cifs-utils 6.2-1 On a second older server (hostname nemesis) I have: 18:30 nemesis:~/scr/mnt> smbd -V Version 3.4.5-5.1-2300-SUSE-SL11.0 kernel-pae-2.6.25.20-0.7 cifs-mount-3.4.5-5.1 Mounting the '/' config share from host *providence on nemesis* works fine: 18:31 nemesis:~/scr/mnt> sudo mount.cifs //providence/config /mnt/pv-cfg -o username=david,uid=1000,credentials=/home/david/.dcr/mountcfile,noperm 18:31 nemesis:~/scr/mnt> l /mnt/pv-cfg total 4 drwxr-xr-x 21 david root 0 2012-09-04 14:41 ./ drwxr-xr-x 14 root root 4096 2013-12-28 18:29 ../ drwxr-xr-x 2 david root 0 2012-09-04 14:50 bin/ drwxr-xr-x 3 david root 0 2012-09-04 14:48 boot/ <snip> Mounting the '/' share from host *nemesis on providence* works fine: 18:37 providence:~/scr/mnt> sudo mount.cifs //nemesis/config /mnt/nm-cfg -o username=david,uid=1000,credentials=/home/david/.dcr/mountcfile,noperm 18:37 providence:~/scr/mnt> l /mnt/nm-cfg total 4 drwxr-xr-x 21 david root 0 Dec 9 11:24 . drwxr-xr-x 11 root root 4096 Mar 3 2011 .. drwxr-xr-x 2 david root 0 Dec 5 2010 bin drwxr-xr-x 4 david root 0 Mar 8 2010 boot <snip> Mounting the '/' share from host *providence on phoinix* works fine: 18:54 phoinix:~/scr/mnt> sudo mount.cifs //providence/config /mnt/pv-cfg -o username=david,uid=1000,credentials=/home/david/.dcr/mountcfile,noperm 18:55 phoinix:~/scr/mnt> l /mnt/pv-cfg/ total 4 drwxr-xr-x 21 david root 0 Sep 4 2012 . drwxr-xr-x 9 root root 4096 Dec 28 18:42 .. drwxr-xr-x 2 david root 0 May 13 2011 .config drwxr-xr-x 2 david root 0 Sep 4 2012 bin drwxr-xr-x 3 david root 0 Sep 4 2012 boot <snip> Mounting the '/' share from host *nemesis on phoinix* works fine: 18:55 phoinix:~/scr/mnt> sudo mount.cifs //nemesis/config /mnt/nm-cfg -o username=david,uid=1000,credentials=/home/david/.dcr/mountcfile,noperm 18:57 phoinix:~/scr/mnt> l /mnt/nm-cfg total 4 drwxr-xr-x 21 david root 0 Dec 9 11:24 . drwxr-xr-x 9 root root 4096 Dec 28 18:42 .. drwxr-xr-x 2 david root 0 Dec 5 2010 bin drwxr-xr-x 4 david root 0 Mar 8 2010 boot <snip> However, attempting the mount of the '/' share from host *phoinix on anything* else fails. The only difference I can see is this is when I'm attempting to mount the share from 4.1.X on anything else: 18:59 nemesis:~/scr/mnt> sudo mount.cifs //phoinix/config /mnt/phx-cfg -o username=david,uid=1000,credentials=/home/david/.dcr/mountcfile,noperm mount error(13): Permission denied Refer to the mount.cifs(8) manual page (e.g. man mount.cifs) 18:40 providence:~/scr/mnt> sudo mount.cifs //phoinix/config /mnt/phx-cfg -o username=david,uid=1000,credentials=/home/david/.dcr/mountcfile,noperm mount error(13): Permission denied Refer to the mount.cifs(8) manual page (e.g. man mount.cifs) The ASCII packet dissection for the STATUS_ACCESS_DENIED packet is: No. Time Source Destination Protocol Length Info 25 3.487933 192.168.7.16 192.168.7.124 SMB 105 Tree Connect AndX Response, Error: STATUS_ACCESS_DENIED Frame 25: 105 bytes on wire (840 bits), 105 bytes captured (840 bits) WTAP_ENCAP: 1 Arrival Time: Jan 6, 2014 17:45:50.826685000 CST [Time shift for this packet: 0.000000000 seconds] Epoch Time: 1389051950.826685000 seconds [Time delta from previous captured frame: 0.001539000 seconds] [Time delta from previous displayed frame: 0.001539000 seconds] [Time since reference or first frame: 3.487933000 seconds] Frame Number: 25 Frame Length: 105 bytes (840 bits) Capture Length: 105 bytes (840 bits) [Frame is marked: False] [Frame is ignored: False] [Protocols in frame: eth:ip:tcp:nbss:smb] [Coloring Rule Name: SMB] [Coloring Rule String: smb || nbss || nbns || nbipx || ipxsap || netbios] Ethernet II, Src: Micro-St_1a:8c:fa (00:21:85:1a:8c:fa), Dst: Dell_22:50:08 (00:11:43:22:50:08) Destination: Dell_22:50:08 (00:11:43:22:50:08) Address: Dell_22:50:08 (00:11:43:22:50:08) .... ..0. .... .... .... .... = LG bit: Globally unique address (factory default) .... ...0 .... .... .... .... = IG bit: Individual address (unicast) Source: Micro-St_1a:8c:fa (00:21:85:1a:8c:fa) Address: Micro-St_1a:8c:fa (00:21:85:1a:8c:fa) .... ..0. .... .... .... .... = LG bit: Globally unique address (factory default) .... ...0 .... .... .... .... = IG bit: Individual address (unicast) Type: IP (0x0800) Internet Protocol Version 4, Src: 192.168.7.16 (192.168.7.16), Dst: 192.168.7.124 (192.168.7.124) Version: 4 Header length: 20 bytes Differentiated Services Field: 0x00 (DSCP 0x00: Default; ECN: 0x00: Not-ECT (Not ECN-Capable Transport)) 0000 00.. = Differentiated Services Codepoint: Default (0x00) .... ..00 = Explicit Congestion Notification: Not-ECT (Not ECN-Capable Transport) (0x00) Total Length: 91 Identification: 0x0398 (920) Flags: 0x02 (Don't Fragment) 0... .... = Reserved bit: Not set .1.. .... = Don't fragment: Set ..0. .... = More fragments: Not set Fragment offset: 0 Time to live: 64 Protocol: TCP (6) Header checksum: 0xa728 [correct] [Good: True] [Bad: False] Source: 192.168.7.16 (192.168.7.16) Destination: 192.168.7.124 (192.168.7.124) [Source GeoIP: Unknown] [Destination GeoIP: Unknown] Transmission Control Protocol, Src Port: microsoft-ds (445), Dst Port: 50813 (50813), Seq: 210, Ack: 445, Len: 39 Source port: microsoft-ds (445) Destination port: 50813 (50813) [Stream index: 2] Sequence number: 210 (relative sequence number) [Next sequence number: 249 (relative sequence number)] Acknowledgment number: 445 (relative ack number) Header length: 32 bytes Flags: 0x018 (PSH, ACK) 000. .... .... = Reserved: Not set ...0 .... .... = Nonce: Not set .... 0... .... = Congestion Window Reduced (CWR): Not set .... .0.. .... = ECN-Echo: Not set .... ..0. .... = Urgent: Not set .... ...1 .... = Acknowledgment: Set .... .... 1... = Push: Set .... .... .0.. = Reset: Not set .... .... ..0. = Syn: Not set .... .... ...0 = Fin: Not set Window size value: 235 [Calculated window size: 30080] [Window size scaling factor: 128] Checksum: 0x5220 [validation disabled] [Good Checksum: False] [Bad Checksum: False] Options: (12 bytes), No-Operation (NOP), No-Operation (NOP), Timestamps No-Operation (NOP) Type: 1 0... .... = Copy on fragmentation: No .00. .... = Class: Control (0) ...0 0001 = Number: No-Operation (NOP) (1) No-Operation (NOP) Type: 1 0... .... = Copy on fragmentation: No .00. .... = Class: Control (0) ...0 0001 = Number: No-Operation (NOP) (1) Timestamps: TSval 290109681, TSecr 203763093 Kind: Timestamp (8) Length: 10 Timestamp value: 290109681 Timestamp echo reply: 203763093 [SEQ/ACK analysis] [This is an ACK to the segment in frame: 24] [The RTT to ACK the segment was: 0.001539000 seconds] [Bytes in flight: 39] NetBIOS Session Service Message Type: Session message (0x00) Length: 35 SMB (Server Message Block Protocol) SMB Header Server Component: SMB [Response to: 24] [Time from request: 0.001539000 seconds] SMB Command: Tree Connect AndX (0x75) NT Status: STATUS_ACCESS_DENIED (0xc0000022) Flags: 0x80 1... .... = Request/Response: Message is a response to the client/redirector .0.. .... = Notify: Notify client only on open ..0. .... = Oplocks: OpLock not requested/granted ...0 .... = Canonicalized Pathnames: Pathnames are not canonicalized .... 0... = Case Sensitivity: Path names are case sensitive .... ..0. = Receive Buffer Posted: Receive buffer has not been posted .... ...0 = Lock and Read: Lock&Read, Write&Unlock are not supported Flags2: 0xc003 1... .... .... .... = Unicode Strings: Strings are Unicode .1.. .... .... .... = Error Code Type: Error codes are NT error codes ..0. .... .... .... = Execute-only Reads: Don't permit reads if execute-only ...0 .... .... .... = Dfs: Don't resolve pathnames with Dfs .... 0... .... .... = Extended Security Negotiation: Extended security negotiation is not supported .... .0.. .... .... = Reparse Path: The request does not use a @GMT reparse path .... .... .0.. .... = Long Names Used: Path names in request are not long file names .... .... ...0 .... = Security Signatures Required: Security signatures are not required .... .... .... 0... = Compressed: Compression is not requested .... .... .... .0.. = Security Signatures: Security signatures are not supported .... .... .... ..1. = Extended Attributes: Extended attributes are supported .... .... .... ...1 = Long Names Allowed: Long file names are allowed in the response Process ID High: 0 Signature: 0000000000000000 Reserved: 0000 Tree ID: 0 (\\phoinix\config) [Path: \\phoinix\config] [Mapped in: 25] Process ID: 1996 User ID: 14822 Multiplex ID: 3 Tree Connect AndX Response (0x75) Word Count (WCT): 0 Byte Count (BCC): 0 -- David C. Rankin, J.D.,P.E.

2 7

[arch-general] Logind Timing out?
by Steve Holmes 12 Jan '14

12 Jan '14

After updating my 64 bigt Arch system to latest stable releases, I began to have problems with logind services timing out and allowing me to have only one console available to work with. it seems that when I have ntpdate running and use netctl to auto start my wireless profile, these problems occur. When I disable ntpdate.service, then I seem to get my consols back and logind appears to be working OK. I also experienced similar problems when I was trying out netctl-ifplugd with my wired connections. If the cable was plugged in, then all started right up just fine but if I unplugged the cable and rebooted, then the above stated problems also wold occur. Up until today, I was not having these problems. I can't seem to pin down the exact cause of these issues but these began to happen after experimenting with a more dynamic network setup. My ultimate goal is to have both netctl-ifplugd and netctl-auto both running and then I can have the flexibility of switching to a cabled connection if necessary and be able to also dynamically pick up whatever WIFI service I come into. I read the Arch wiki pages about networking so I have most of that working presuably OK now except for these logind issues. How could networking delays cause all the timing problems with logind.service?

1 0

[arch-general] "os-prober" failed to work with "blkid" (util-linux 2.24-2)
by Ken CC 09 Jan '14

09 Jan '14

Hi, It is because the new blkid can now recognize the extened partion, but os-prober relies on the return value "2" (exit_not_found) of blkid to skip processing the extended partition. # fdisk -l ... Device Boot Start End Blocks Id System /dev/sda1 * 63 252847034 126423486 7 HPFS/NTFS/exFAT /dev/sda2 252847035 625137344 186145155 5 Extended /dev/sda5 252847098 253425374 289138+ 83 Linux /dev/sda6 385254828 482913899 48829536 83 Linux ... blkid 2.24-2 --- # blkid /dev/sda2 /dev/sda2: PTTYPE="dos" PARTUUID="28b48ca1-02" # echo $? 0 # blkid -o value -s TYPE /dev/sda2 # no output # echo $? 0 blkid 2.20.0 --- # blkid /dev/sda2 # no output # echo $? 2 # blkid -o value -s TYPE /dev/sda2 # no output # echo $? 2 Affected os-prober code: --- File: /usr/lib/os-probes/50mounted-tests - 8 types="$(fs_type "$partition")" || types=NOT-DETECTED 9 if [ "$types" = NOT-DETECTED ]; then 10 debug "$1 type not recognised; skipping" 11 exit 0 With old blkid, os-prober will just skip here when handling /dev/sda2, but with the current version of blkid, os-prober will go on and try to mount /dev/sda2 with FS types found in the system, which causes lots of errors, even kernel calltrace when trying xfs, on my machine. I think there are three ways to sovlve this: 1. change the way that os-prober detects devices. 2. make blkid return an error code when # blkid -o value -s TYPE /dev/sdaX returns no proper result. 3. use the old version of blkid (util-linux). What do you guys think? -ken

1 0

[arch-general] Thank you
by Genes Lists 09 Jan '14

09 Jan '14

Over the last 2 weeks I have replaced 3 fedora servers with Arch - and want to say thank you . That's the last of my fedora machines now and I am all Arch and they are all working really well. The new boxes are haswell UEFI and boot using refind. Setting this up was straightforward with Arch. Thank you for making Arch the best linux distro. Happy New Year gene

16 16

[arch-general] Package management
by Temlin Olivér 07 Jan '14

07 Jan '14

On Sun, Jan 5, 2014 at 3:07 PM, Kalrish Bäakjen <kalrish.antrax(a)gmail.com> wrote: > > Hello, > > Before starting, I'd like to remind that we're all humans, so we can > discuss these ideas politely and making use of nothing more than > reason. I am NOT a pedantic person looking to impose my will; please, > excuse me if you feel I'm doing it. I hope this turns into an > interesting conversation, and not into an unjustified battle, as it > too often does. > > Recently, I realized I was not using Arch in the way it was > "intended"; I was building most of the packages of my system, much "a > la Gentoo", with modified PKGBUILDs and a local repository. While it > worked with no major inconveniences, I started looking for truly > source-based distros, and found Gentoo. However, I disliked some parts > about it. This is my attempt at taking what I consider to be "good" > ideas to Arch. > > **Binary packages are right** > It is possible to either build packages by yourself, and then store > them for later use, or directly take them in prebuilt form. This is > specially useful when time matters or when the machine is unable to > build things (in my case, I broke a laptop, and it seems it was > because of overheating). > > **Versioning** > If I have understood it right, Gentoo makes it possible to have > several versions of the same package through "slots". Consider the > following scenario in Arch: > bar1 -> libfoo-1.0 > bar2 -> libfoo-1.0 > `libfoo´ is updated to 2.0. `bar2´ catchs up, but `bar1´ does not. > Thus, updating `libfoo´ to the latest version would break `bar1´, but > is necessary for `bar2´. The solution I've seen is adding the major > version number in the package name (e.g. "gtk2", "freetype2"), which > would turn the older version of `libfoo´ into `libfoo1´. I think, > however, that we could implement Gentoo's "slots", and add the concept > of "alternatives". That way, there would be PKGBUILDs for both > versions of the library, each with the same name but different > version. Each library would be installed as, say, > "/usr/lib/libfoo-${pkgver}.so", and an additional script for handling > of the named "alternatives" could then create a symlink at > /usr/lib/libfoo.so pointing to the right version _if needed_. The > script would exist in a per-package basis, so there would only be one > for both versions of `libfoo´. For example: > $ alternatives-manager libfoo --choose 2.0 > ln -s /usr/lib/libfoo-2.0.so /usr/lib/libfoo.so > (The name and syntax is just an example.) > This is easier to understand with libraries. With programs, however, > it's more difficult; someone could want to install different versions > of the same program, or, more likely, an outdated version (a typical > case would be GNOME 2). We could add "-${pkgver}" to the name of the > binaries in such cases, and then use alternatives to create convenient > symlinks. It would be horrible to have "cp-8.9.2" and the like. > Packages that could benefit from this are freetype, python, and I > guess every library out there. It would be, however, hard to > implement. > > **Package selection** > When pacman finds several versions of the same package in various > repositories, it fetches the one in the prevalent repo. This implies > that, if I've configured my local repository before the official > repos, I will never notice that a package has been updated. I'll have > to do `pacman -Ss pkg-that-ive-compiled´ to check if there's a newer > version, or track each project by myself. I think that, if pacman > finds itself in the described situation, it should prompt to ask which > package to install, unless --noconfirm is passed, in which case it > should stick with the default. [If I do `pacman -S repo/pkg´, however, > pacman should have no problem in choosing the package, and it > shouldn't ask.] > > These were all the ideas. If you haven't understood something, please > ask. Thanks, > Kalrish Hello! It's a nice idea to have, but packages need to link to the proper library version (eg. libfoo.so.1, not libfoo.so), thus making the repeated linking unnecesary. If that's a problem, then you can file a bug report. There already are a lot of symlinks in /lib, to ensure compatibility. If a new library version brakes old packages (many), than the old library is released as you said, with corrected dependencies for all packages (see libpng12 in community). Handling old versions is nearly impossible, as neither packagers nor upstream can be bothered to provide support for outdated packages. Even if Gentoo found a way, the biggest problem would be that it goes against the Arch Way: outdated packages go to the repos and it would complicate the system even more, using the libpng example, the AUR contains three additional packages, making the count 5. Putting these in the official repos would encourage developers to keep using old libraries for their projects and force the user to (in your alternatives-manager-style version) to change library symlinks repeatedly. So the arch solution to your problem is to notify bar1 upstream, and if they don't make a solution (usually because debian/ubuntu not having the new library versions yet), then make a custom patch, or if there are only a few users, and/or a dead upstream, drop the package, which has a (high) chance to resurrect in the AUR. Just bringing up an example search for gnome at the forums around 2011-04 and a few months forth. Dozens of people wanted back the old Gnome 2 (including myself), whined (not me), but to no avail. All the answers were all no. Arch is not going to support (or even ship) packages that are out-of-date. The current solution is to use the since-developed MATE and Cinnamon derivatives. But I'm just a user, and these are my thoughts, not completely matching the arch developers' ones. -- Temlin Olivér

4 5

[arch-general] Package management
by Kalrish Bäakjen 06 Jan '14

06 Jan '14

Hello, Before starting, I'd like to remind that we're all humans, so we can discuss these ideas politely and making use of nothing more than reason. I am NOT a pedantic person looking to impose my will; please, excuse me if you feel I'm doing it. I hope this turns into an interesting conversation, and not into an unjustified battle, as it too often does. Recently, I realized I was not using Arch in the way it was "intended"; I was building most of the packages of my system, much "a la Gentoo", with modified PKGBUILDs and a local repository. While it worked with no major inconveniences, I started looking for truly source-based distros, and found Gentoo. However, I disliked some parts about it. This is my attempt at taking what I consider to be "good" ideas to Arch. **Binary packages are right** It is possible to either build packages by yourself, and then store them for later use, or directly take them in prebuilt form. This is specially useful when time matters or when the machine is unable to build things (in my case, I broke a laptop, and it seems it was because of overheating). **Versioning** If I have understood it right, Gentoo makes it possible to have several versions of the same package through "slots". Consider the following scenario in Arch: bar1 -> libfoo-1.0 bar2 -> libfoo-1.0 `libfoo´ is updated to 2.0. `bar2´ catchs up, but `bar1´ does not. Thus, updating `libfoo´ to the latest version would break `bar1´, but is necessary for `bar2´. The solution I've seen is adding the major version number in the package name (e.g. "gtk2", "freetype2"), which would turn the older version of `libfoo´ into `libfoo1´. I think, however, that we could implement Gentoo's "slots", and add the concept of "alternatives". That way, there would be PKGBUILDs for both versions of the library, each with the same name but different version. Each library would be installed as, say, "/usr/lib/libfoo-${pkgver}.so", and an additional script for handling of the named "alternatives" could then create a symlink at /usr/lib/libfoo.so pointing to the right version _if needed_. The script would exist in a per-package basis, so there would only be one for both versions of `libfoo´. For example: $ alternatives-manager libfoo --choose 2.0 ln -s /usr/lib/libfoo-2.0.so /usr/lib/libfoo.so (The name and syntax is just an example.) This is easier to understand with libraries. With programs, however, it's more difficult; someone could want to install different versions of the same program, or, more likely, an outdated version (a typical case would be GNOME 2). We could add "-${pkgver}" to the name of the binaries in such cases, and then use alternatives to create convenient symlinks. It would be horrible to have "cp-8.9.2" and the like. Packages that could benefit from this are freetype, python, and I guess every library out there. It would be, however, hard to implement. **Package selection** When pacman finds several versions of the same package in various repositories, it fetches the one in the prevalent repo. This implies that, if I've configured my local repository before the official repos, I will never notice that a package has been updated. I'll have to do `pacman -Ss pkg-that-ive-compiled´ to check if there's a newer version, or track each project by myself. I think that, if pacman finds itself in the described situation, it should prompt to ask which package to install, unless --noconfirm is passed, in which case it should stick with the default. [If I do `pacman -S repo/pkg´, however, pacman should have no problem in choosing the package, and it shouldn't ask.] These were all the ideas. If you haven't understood something, please ask. Thanks, Kalrish

3 4

[arch-general] Default value of "j" in makeflags of makepkg.conf
by Mark Lee 06 Jan '14

06 Jan '14

Salutations, I was wondering if we could change the default value for j in MakeFlags in makepkg.conf to "-j$(nproc)". This would allow makepkg to scale the number of threads per pc as default. Regards, Mark -- Mark Lee <mark(a)markelee.com>

15 33

[arch-general] MAKEFLAGS remodelation
by Kalrish Bäakjen 06 Jan '14

06 Jan '14

Hello, Some projects (e.g., mpv, in community) are moving from Autotools to other build systems. I think we should create a JOBS variable indicating the number of parallel jobs, leaving MAKEFLAGS for make-specific options. We could also implement SCONSFLAGS, WAFFLAGS,… although that may be a bit premature. The reason behind this is that a given build system might not understand "-j N". (And, also, for SCons projects, the "-Q" flag could be desired.) In the proposed model, each PKGBUILD would appropiately tell the build system how many jobs to do in parallel, or ignore the setting. As an extra question, may I ask why aren't MAKEFLAGS used in official PKGBUILDs? I mean, as in: make ${MAKEFLAGS} Perhaps make automatically fetches options from the environment? Or maybe because package construction is automated in a server and other processes are running in other cores? Thanks, Kalrish

4 9

Re: [arch-general] Default value of "j" in makeflags of makepkg.conf
by Isaac Dupree 04 Jan '14

04 Jan '14

On 01/03/2014 04:03 PM, Mark Lee wrote: > On Fri, 2014-01-03 at 15:55 -0500, Isaac Dupree wrote: >> On 01/03/2014 10:37 AM, Anatol Pomozov wrote: >>> Using -j$(cpunum) is a sane default that saves a lot of time to users. >> >> I agree, but for the record, 'nice' and scheduling are no panacea in my >> experience. It's fine for CPU loads, but compilations are also >> disk-heavy (which mattered when I used a spinning disk) and sometimes >> RAM-heavy (-j8 on my pet C++ project uses 2GB of RAM, which could push >> other programs I'm using into swap). >> >> -Isaac >> >> > > Salutations, > > I got to ask why you are running make with 8 threads on a system with > only 2 GB of ram? Heh. Well, I have 8 GB of RAM, but I'm often using most of it already. $(nproc) is 8 because there are 4 cores with each with hyperthreading (Intel Core i7). The project builds roughly at the same speed on this box at anywhere from -j4 to -j8; if that generalizes to other projects and there's an easy way to count cores not hyperthreads, cores might be a better Arch default. Shrugs. There are quite a few 4GB i7 systems out there too. Quad core ARM systems are another point on the cores-to-RAM-ratio spectrum. Probably most code doesn't consume 256 MB RAM per GCC invocation. C++'s templates and lack of module system... -Isaac

2 1

[arch-general] USB S-ATA Adapter
by Silvio Siefke 03 Jan '14

03 Jan '14

Hello, my Notebook is broken and so i buy a Adapter that i can take my Data from the Harddisk. siefke ~ $ lsusb Bus 001 Device 003: ID 0402:9665 ALi Corp. Gateway Webcam Bus 001 Device 004: ID 174c:5106 ASMedia Technology Inc. Transcend StoreJet 25M3 siefke ~ $ ls /dev | grep sd sda sda1 sda2 sdb SDB is the HDD but i can not mount and fdisk give nothing out. siefke ~ $ sudo fdisk -l [sudo] password for siefke: Disk /dev/sda: 232,9 GiB, 250059350016 bytes, 488397168 sectors Units: Sektoren of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disklabel type: dos Disk identifier: 0x47a0cee0 Gerät Boot Start Ende Blocks Id System /dev/sda1 2048 4196351 2097152 82 Linux swap / Solaris /dev/sda2 * 4196352 488397167 242100408 83 Linux Has someone an idea or can i forget it? Thank you for help & Nice Day Silvio

5 4