Re: The rant about browsers
> ok, how do I put this nicely... > To run a modern browser, you need a modern computer. 1.5GB RAM and a > celeron processor doesn't cut it. > Nick Moving towards a "modern" computers one will have problems with supported hardware. Maybe some desktops are ok, but what can you do about laptops. There is no documentation and manufactures are pushing all kinds of crazy shits like fake RAID, UEFI, ACPI, etc. New software is pressing for new hardware too. There is a stupid movement for browsers too: W3C approves and is trying to make a strandard for any shit you can bring inside a browser. Developers are following - who would want users to leave because "X" browser doesn't support "Y" feature. This rush has only one 'benefit', more money for harware manufacturers' pockets. I let the people with more experience to anticipate and describe the future. Thanks.
Re: amavisd uses high cpu usage?
> Hi Stuart, > > > I think may be something is wrong with perl modules. > > Could be. How did you install things? Is this a fresh 5.5 install or an > upgrade from an earlier version? Did you upgrade all packages? > Yeah, This is a fresh 5.5 64 bit. I downloaded iso from this http://ftp.jaist.ac.jp/pub/OpenBSD/5.5/amd64/install55.iso and burnt it to CD and installed. Then, I downloaded below stuffs and copied to a CD . http://ftp.jaist.ac.jp/pub/OpenBSD/5.5/src.tar.gz http://ftp.jaist.ac.jp/pub/OpenBSD/5.5/sys.tar.gz http://ftp.jaist.ac.jp/pub/OpenBSD/5.5/ports.tar.gz Then I mounted the cd and copied them to /usr directory and ran below commands as explained here. http://www.openbsd.org/anoncvs.html#starting # *cd /usr/src* # *tar xzf ../sys.tar.gz* # *tar xzf ../src.tar.gz* # *cd /usr* # *tar xzf ports.tar.gz* But, I still have NOT updated src and port tree. Could you pls let me kow which command do I need? I want to stay in Stable branch - it is patched branch isn't it? Are below commands ok? to update the src # cd /usr/src Now which is the right command for a patch branch? cvs -d anon...@anoncvs.jp.openbsd.org:/cvs -q up -Pd or cvs -d anon...@anoncvs.jp.openbsd.org:/cvs -q up -rOPENBSD_5_5 -Pd to update the port tree. # cd /usr/ports Again, which is the right command for a patch branch? cvs -d anon...@anoncvs.jp.openbsd.org:/cvs -q up -Pd or cvs -d anon...@anoncvs.jp.openbsd.org:/cvs -q up -rOPENBSD_5_5 -Pd Pls let me know. I will run and add a cronjob on daily basis. Anyway, I installed posfix amavisd-new with pkg_add -v command. export PKG_PATH=http://ftp.jaist.ac.jp/pub/OpenBSD/5.5/packages/amd64/ then pkg_add -v postfix pkg_add -v amavisd-new potput of pkg_info amavisd-new-2.8.1p0 interface between mailer MTA and content checkers arc-5.21p create & extract files from DOS .ARC files autoconf-2.13p3 automatically configure source code on many Un*x platforms autoconf-2.65p0 automatically configure source code on many Un*x platforms autoconf-2.67p0 automatically configure source code on many Un*x platforms autoconf-2.68p0 automatically configure source code on many Un*x platforms autoconf-2.69p1 automatically configure source code on many Un*x platforms automake-1.10.3p7 GNU Standards-compliant Makefile generator automake-1.11.6p1 GNU Standards-compliant Makefile generator automake-1.12.6p0 GNU Standards-compliant Makefile generator avahi-0.6.31p13 framework for Multicast DNS Service Discovery bash-4.2.45p0 GNU Bourne Again Shell bison-2.3p1 GNU parser generator bzip2-1.0.6p0 block-sorting file compressor, unencumbered cabextract-1.4 extracts files from Microsoft CAB archives cairo-1.12.16 vector graphics library clamav-0.98.1 virus scanner cups-libs-1.7.1 CUPS libraries and headers curl-7.34.0p0 get files from FTP, Gopher, HTTP or HTTPS servers cyrus-sasl-2.1.26p10 RFC SASL (Simple Authentication and Security Layer) db-4.6.21p0v0 Berkeley DB package, revision 4 dbus-1.8.0v0message bus system docbook-4.5p1 technical documentation XML/SGML definitions docbook-dsssl-1.79 modular DSSSL stylesheets for the DocBook DTD docbook-xsl-1.68.1p5 docbook XSL modular stylesheet e2fsprogs-1.42.7p0 utilities to manipulate ext2 filesystems easy-rsa-2.2.0p0small RSA key management package gd-2.0.35p1 library for dynamic creation of images gdbm-1.10p0 GNU dbm gettext-0.18.2p4GNU gettext glib2-2.38.2p6 general-purpose utility library gmake-4.0p0 GNU make gnugetopt-1.1.4p2 GNU getopt(1) utility gnupg-1.4.16GNU privacy guard - a free PGP replacement gobject-introspection-1.38.0p1 GObject Introspection gperf-3.0.4p0 perfect hash functions, to help write parsers groff-1.22.2p4 GNU troff typesetter help2man-1.41.1p0 generates simple manual pages from program output icu4c-52.1 International Components for Unicode iftop-1.0pre2p0 display bandwidth usage on an interface intltool-0.50.2 internationalization tools iso8879-1986p0 character entity sets from ISO 8879:1986 (SGML) jdk-1.6.0.32p6 Java2(TM) SE Dev Kit v1.6.0.32 jnettop-0.13.0p1capture network traffic, display streams sorted by bandwidth jpeg-9p0IJG's JPEG compression utilities lha-1.14i.ac20050924.1 archive files using LZW compression (.lzh files) libdaemon-0.14p0lightweight C library that eases the writing of daemons libelf-0.8.13p1 read, modify, create ELF files on any arch libexecinfo-0.2p3v0 clone of backtrace facility found in the GNU libc libffi-3.0.9p6 Foreign Function Interface libgcrypt-1.6.1 crypto library based on code used in GnuPG libghttp-1.0.9p2GNOME http client library libgpg-error-1.12p0 error codes for GnuPG related software libiconv-1.14p1 character set conversion library libidn-1.28p0 internationalized string handling libltdl-2.4.2p0 GNU
Re: The rant about browsers
On 23.08.2014. 18:16, Nick Holland wrote: real mem = 1568260096 (1495MB) avail mem = 1517772800 (1447MB) ... cpu0 at mainbus0: apid 0 (boot processor) cpu0: Intel(R) Celeron(R) CPU G530 @ 2.40GHz, 2394.94 MHz ok, how do I put this nicely... To run a modern browser, you need a modern computer. 1.5GB RAM and a celeron processor doesn't cut it. NOW, that doesn't cause CRASHES, but when you fix the crashes by cranking up your login.conf specs, you will be so far into swap you will wish your browser crashed. Well, nowadays one can get a very fast CPU and lot of RAM cheaply, but that does not mean all of this is necessary in order to just browse the web. From time to time, I must use a 12-year old Pentium 4 Northwood, 1.8 GHz, 512 kB cache with 512 MB RAM, it has Windows XP installed and is quite usable with modern web browsers. Until recently, I also regularly used an Athlon64 Venice, 2.0 GHz, 1 MB cache and 1 GB RAM under Linux, and it was usable even with many tabs/sites open. The only problem was Adobe flash Linux plugin, which was for some reason slower than its Windows counterpart. Current Pentiums and Celerons (such as this G530) are based on Core i architecture, have more than one core, and are much faster than the two mentioned processors. 1.5 GB RAM is also _a lot of memory_, regardless how easy is to get more today. The point is: It should work just fine. Just raise the OS memory limits.
Re: Dell PE2900 instant-reboot with 5.5-RELEASE
On 08/23/14 18:30, Adam Thompson wrote: On 14-08-23 05:49 PM, Adam Thompson wrote: Copying install55.fs to a USB stick and booting from it starts to boot, gets part-way through the boot process, then suddenly reboots. All amd64 images fail in exactly the same way. The server logs a Machine Check Exception on CPU1 along with a bunch of "Unknown OEM Sensor triggered" events. (It's a Dell - how can the OEM [i.e. Dell!] sensors be unknown???) And now begins the torturous process of determining what changed - booting a 4.0 amd64 CD works just fine. I was about to start a binary search to find the last version that worked correctly, but I see the main site and mirrors all go back as far as 5.3 and no further. My google-fu is weak, apparently - where do I find (i.e. download) older releases that I can't find my CDs for? I would try a CD image to see if it reboots. Use the DRAC, if available, to log the output. I have seen bad memory cause machine check exceptions. Although it's odd that an older OpenBSD image boots. Might want to run Dell diags to see if it turns up anything. Or reseat all of the DIMMs to see if that helps. The PE2900 is a bit old. A little more modern hardware, if possible, might be a better option. One of my favourite Dell machines is the R900, but it's pretty old too... Stan
Re: Dell PE2900 instant-reboot with 5.5-RELEASE
On 08/23/14 18:49, Adam Thompson wrote: > Posting here before filing a bug in case this rings a bell... > > Dell PowerEdge 2900, with PERC 5 "integrated" controller. All BIOSes > and firmware levels up to date as of ~6 months ago. (Which should be > pretty current, since this isn't a new system!) > > Copying install55.fs to a USB stick and booting from it starts to boot, > gets part-way through the boot process, then suddenly reboots. This machine has a CDROM drive on it, and Dell has had some fantasticly buggy USB support in the BIOS in the past. Use the CDROM. Or the ISO file through the DRAC. I know I've booted some OSs from USB drives on 2950s, but I don't recall if OpenBSD was one of them. > This server does the exact same thing with Ubuntu 12.04.x, but otherwise > works fine: FreeBSD (via FreeNAS), DragonflyBSD 3.8.2, Debian (via > ProxmoxVE), CentOS 6.4 all boot and work just fine; it's been operating > as a Proxmox server in production for the last 12 months without any > issues, so I'm fairly confident it's not a hardware problem. > > I don't even know where to start with this... ideas? Right now, I'm > re-writing the install FS so I can re-test, then I'll try an actual > CD-ROM. I can take video of the boot screen, not sure how to get serial > console output that early in the process. If your machine won't boot a CDROM image, I suspect your machine is broke. While I've never had a Dell PE2900 in my hands, the PE2950 is supposedly a very similar machine...and I think it is very very safe to say that OpenBSD works wonderfully on 2950s, and I'm pretty sure I've loaded Ubuntu 12.04 on 'em, as well. Nick.
Re: Dell PE2900 instant-reboot with 5.5-RELEASE
On 14-08-23 05:49 PM, Adam Thompson wrote: Copying install55.fs to a USB stick and booting from it starts to boot, gets part-way through the boot process, then suddenly reboots. All amd64 images fail in exactly the same way. The server logs a Machine Check Exception on CPU1 along with a bunch of "Unknown OEM Sensor triggered" events. (It's a Dell - how can the OEM [i.e. Dell!] sensors be unknown???) And now begins the torturous process of determining what changed - booting a 4.0 amd64 CD works just fine. I was about to start a binary search to find the last version that worked correctly, but I see the main site and mirrors all go back as far as 5.3 and no further. My google-fu is weak, apparently - where do I find (i.e. download) older releases that I can't find my CDs for? -- -Adam Thompson athom...@athompso.net
Dell PE2900 instant-reboot with 5.5-RELEASE
Posting here before filing a bug in case this rings a bell... Dell PowerEdge 2900, with PERC 5 "integrated" controller. All BIOSes and firmware levels up to date as of ~6 months ago. (Which should be pretty current, since this isn't a new system!) Copying install55.fs to a USB stick and booting from it starts to boot, gets part-way through the boot process, then suddenly reboots. This server does the exact same thing with Ubuntu 12.04.x, but otherwise works fine: FreeBSD (via FreeNAS), DragonflyBSD 3.8.2, Debian (via ProxmoxVE), CentOS 6.4 all boot and work just fine; it's been operating as a Proxmox server in production for the last 12 months without any issues, so I'm fairly confident it's not a hardware problem. I don't even know where to start with this... ideas? Right now, I'm re-writing the install FS so I can re-test, then I'll try an actual CD-ROM. I can take video of the boot screen, not sure how to get serial console output that early in the process. -- -Adam Thompson athom...@athompso.net
Re: OpenBSD 5.5-STABLE: Full Disk Encryption (bioctl) and Smard Cards
> > >Are there any YubiKey-like devices that can contain many static > > >password, not one like YubiKey? > > > > Not sure it helps, but mine contains two... > > It helps! I need one for login password and second for firefox's password > manager. Which model do you use? All yubikeys have the two slots, to my knowledge, which can be set either static or otp.
Re: The rant about browsers
On Sat, Aug 23, 2014 at 08:15:23PM +0200, Peter J. Philipp wrote: > However I have a different problem. I use firefox over ssh to another > user on the same system. I do this because I don't want a would-be > attacker to get to sensitive files such as my ssh keys. Now this setup > runs pretty good, except at one point and perhaps someone can look into > this for me. When I control-f for searching a website and enter 3 > characters the browser crashes. However it doesn't happen always and it > never happens when I run firefox as my own user. works for me. > Another drawback to my using another user to sandbox firefox is that I > cannot copy-paste from browser to another window, not sure if that is > related. just select the text, and paste it with the third mouse button (usually the mouse wheel) > Anyhow for memory I'm set with 32 GB so that's not the problem in this > system. 32 GB RAM? Not bad... berger s.
Re: The rant about browsers
On 08/23/14 19:59, Amit Kulkarni wrote: > That is your problem...memory You will definitely see better performance > with more memory. I use Pentium G2020 with 8GB of memory and the > performance is good for browsing/occasional video with daily restart. Tweak > the follwoing variables in /etc/login.conf > > datasize-max === 3G > datasize-cur === 2G > I'm going to say something but not sure if it would be seen as a hijacking of the thread, if so, let me know and I'll take it to another thread. I use firefox too and I have never adjusted my datasize yet, never needed too. There is only a few websites that crash it and I don't usually visit those. However I have a different problem. I use firefox over ssh to another user on the same system. I do this because I don't want a would-be attacker to get to sensitive files such as my ssh keys. Now this setup runs pretty good, except at one point and perhaps someone can look into this for me. When I control-f for searching a website and enter 3 characters the browser crashes. However it doesn't happen always and it never happens when I run firefox as my own user. Another drawback to my using another user to sandbox firefox is that I cannot copy-paste from browser to another window, not sure if that is related. Anyhow for memory I'm set with 32 GB so that's not the problem in this system. Sincerely, -peter
Re: The rant about browsers
On Sat, Aug 23, 2014 at 11:16 AM, Nick Holland wrote: > On 08/23/14 10:30, Gregory Edigarov wrote: > > Hello Everybody. > > > > Before anything I want to say big thanks to the developers of OpenBSD, > > for maintaining it. After some ~10 years of being the loyal OpenBSD > > user, I never had any problem with OpenBSD itself, besides may be 2 or > > three times. > > It is impressive. Every other system I use gives problems from time to > > time, so I am thanking you, guys, every time I type a command. > > > > Now onto the bitter part. For some reason, since, may be, AFAIR 5.2 > > times, I do not see any browser that is working flawlessly under our > > loved system. > > Everything is happened on the same set of sites I use routinely everyday. > > > > I tried: > > Firefox - bad, bad, bad. It fails 1000 times a day. > > On your machine, firefox couldn't be restarted 1000 times a day. > (ok, not sure where my sense if irony is today...) > ... > > > dmesg follows: > > OpenBSD 5.6-current (GENERIC.MP) #340: Fri Aug 22 15:06:09 MDT 2014 > > dera...@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC.MP > > real mem = 1568260096 (1495MB) > > avail mem = 1517772800 (1447MB) > ... > > cpu0 at mainbus0: apid 0 (boot processor) > > cpu0: Intel(R) Celeron(R) CPU G530 @ 2.40GHz, 2394.94 MHz > > ok, how do I put this nicely... > To run a modern browser, you need a modern computer. 1.5GB RAM and a > celeron processor doesn't cut it. > NOW, that doesn't cause CRASHES, but when you fix the crashes by > cranking up your login.conf specs, you will be so far into swap you will > wish your browser crashed. > > Modern browsers leak memory like everyone has 16GB and a quad-core proc, > AND restarts their browser several times a day. Look at those same > browsers on Windows (their target market), you see the same thing. The > difference is, OpenBSD kicks out programs that exceed predefined limits, > that's what you are most likely seeing. > > But most likely, login.conf will fix your crash problem, as I use > firefox, Chromium and Thunderbird on my amd64 system (three-core, 4G > RAM), and usually get a week or two uptime between shutdowns (because of > hitting RAM limits). > > Nick. > +1 That is your problem...memory You will definitely see better performance with more memory. I use Pentium G2020 with 8GB of memory and the performance is good for browsing/occasional video with daily restart. Tweak the follwoing variables in /etc/login.conf datasize-max === 3G datasize-cur === 2G
Re: OpenBSD 5.5-STABLE: Full Disk Encryption (bioctl) and Smard Cards
On Sat, Aug 23, 2014 at 02:09:20PM +0200, Alexander Hall wrote: > > > On August 23, 2014 4:33:55 AM CEST, "Артур Истомин" > wrote: > >On Fri, Aug 22, 2014 at 04:03:59PM -0700, Zach Leslie wrote: > >> > However, I don't know how it is seen by the system and if it would > >> > show up as a drive. Anyone in here is using a smart card to decrypt > >> > volumes at boot? > >> > >> You could use a YubiKey with a static long password to unlock the > >boot > >> volume. > > > >[offtop] > > > >Are there any YubiKey-like devices that can contain many static > >password, not one like YubiKey? > > Not sure it helps, but mine contains two... It helps! I need one for login password and second for firefox's password manager. Which model do you use?
Re: New queueing system and HZ value limits
Henning Brauer [hb-open...@ml.bsws.de] wrote: > > Any idea why this was so much less of a problem with altq? > > it wasn't... the hfsc core was the same, and cbq worked exactly the same > way too. > > People might not have paid as much attention? I dunno. > Raising HZ was frowned upon when I ported altq because it sped up _everything_ for the benefit of a potentially unused subsystem. I bet there is a technique to be learned from tickless kernels here.
Re: The rant about browsers
23.8.2014 17:31, Gregory Edigarov kirjoitti: Hello Everybody. Before anything I want to say big thanks to the developers of OpenBSD, for maintaining it. After some ~10 years of being the loyal OpenBSD user, I never had any problem with OpenBSD itself, besides may be 2 or three times. It is impressive. Every other system I use gives problems from time to time, so I am thanking you, guys, every time I type a command. Now onto the bitter part. For some reason, since, may be, AFAIR 5.2 times, I do not see any browser that is working flawlessly under our loved system. Everything is happened on the same set of sites I use routinely everyday. I tried: Firefox - bad, bad, bad. It fails 1000 times a day. Chromium - it is better, in terms. Yes, it will not fail on the plain place (it is a Russian idiom, which means 'from nothing' or 'from no reason one can observe'), but left for some time it starts to be so slow... was forced to stay away from it too. but after all it is the only browser under OpenBSD that have a working lastpass plugin. (and I need lastpass, if I want to share my passwords between home and job computers) Seamonkey - potentially good project. but suffers from the same problems like firefox. although it is fails much much less, the frequency is still unacceptable for me. I also used xombrero and it was good, but again, from somewhere between 5.2 - 5.3 times it has started to fail with an unacceptable frequency. I know, I should write to upstream mailing lists of the projects I've mentioned above, but before that, I want to know if somebody else is suffering such problems and I am still sure maintatiners of the corresponding ports will do it better than me if they find it is a problem. -- With best regards, Gregory Edigarov dmesg follows: OpenBSD 5.6-current (GENERIC.MP) #340: Fri Aug 22 15:06:09 MDT 2014 dera...@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC.MP real mem = 1568260096 (1495MB) avail mem = 1517772800 (1447MB) mpath0 at root scsibus0 at mpath0: 256 targets mainbus0 at root bios0 at mainbus0: SMBIOS rev. 2.6 @ 0xeb170 (91 entries) bios0: vendor American Megatrends Inc. version "0701" date 07/04/2012 bios0: ASUSTeK COMPUTER INC. P8H61-M2 USB3 acpi0 at bios0: rev 2 acpi0: sleep states S0 S1 S3 S4 S5 acpi0: tables DSDT FACP APIC SSDT MCFG HPET acpi0: wakeup devices PS2K(S4) PS2M(S4) BR20(S3) EUSB(S4) USBE(S4) PEX0(S4) PEX1(S4) PEX3(S4) PEX5(S4) PEX6(S4) PEX7(S4) P0P1(S4) P0P2(S4) P0P3(S4) P0P4(S4) acpitimer0 at acpi0: 3579545 Hz, 24 bits acpimadt0 at acpi0 addr 0xfee0: PC-AT compat cpu0 at mainbus0: apid 0 (boot processor) cpu0: Intel(R) Celeron(R) CPU G530 @ 2.40GHz, 2394.94 MHz cpu0: FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,POPCNT,DEADLINE,XSAVE,NXE,LONG,LAHF,PERF,ITSC cpu0: 256KB 64b/line 8-way L2 cache cpu0: smt 0, core 0, package 0 mtrr: Pentium Pro MTRR support, 10 var ranges, 88 fixed ranges cpu0: apic clock running at 99MHz cpu0: mwait min=64, max=64, C-substates=0.2.1.1.0, IBE cpu1 at mainbus0: apid 2 (application processor) cpu1: Intel(R) Celeron(R) CPU G530 @ 2.40GHz, 2394.57 MHz cpu1: FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,POPCNT,DEADLINE,XSAVE,NXE,LONG,LAHF,PERF,ITSC cpu1: 256KB 64b/line 8-way L2 cache cpu1: smt 0, core 1, package 0 ioapic0 at mainbus0: apid 0 pa 0xfec0, version 20, 24 pins acpimcfg0 at acpi0 addr 0xe000, bus 0-63 acpihpet0 at acpi0: 14318179 Hz acpiprt0 at acpi0: bus 0 (PCI0) acpiprt1 at acpi0: bus 2 (PEX0) acpiprt2 at acpi0: bus 3 (PEX1) acpiprt3 at acpi0: bus 5 (PEX3) acpiprt4 at acpi0: bus 6 (PEX5) acpiprt5 at acpi0: bus -1 (PEX6) acpiprt6 at acpi0: bus -1 (PEX7) acpiprt7 at acpi0: bus 1 (P0P1) acpiprt8 at acpi0: bus -1 (P0P2) acpiprt9 at acpi0: bus -1 (P0P3) acpiprt10 at acpi0: bus -1 (P0P4) acpicpu0 at acpi0: C3, C2, C1, PSS acpicpu1 at acpi0: C3, C2, C1, PSS acpibtn0 at acpi0: PWRB acpivideo0 at acpi0: GFX0 acpivout0 at acpivideo0: DD02 cpu0: Enhanced SpeedStep 2394 MHz: speeds: 2400, 2300, 2200, 2100, 2000, 1900, 1800, 1700, 1600 MHz pci0 at mainbus0 bus 0 pchb0 at pci0 dev 0 function 0 "Intel Core 2G Host" rev 0x09 ppb0 at pci0 dev 1 function 0 "Intel Core 2G PCIE" rev 0x09: msi pci1 at ppb0 bus 1 vga1 at pci0 dev 2 function 0 "Intel HD Graphics 2000" rev 0x09 intagp at vga1 not configured inteldrm0 at vga1 drm0 at inteldrm0 drm: Memory usable by graphics device = 2048M inteldrm0: 1280x1024 wsdisplay0 at vga1 mux 1: console (std, vt100 emulation) wsdisplay0: screen 1-5 added (std, vt100 emulation) "Intel 6 Series MEI" rev 0x04 at pci0 dev 22 function 0 not configured ehci0 at pci0 dev 26 function 0 "Intel 6 Series USB" rev 0x05: apic 0 int 23 usb0 at
Re: The rant about browsers
On 08/23/14 10:30, Gregory Edigarov wrote: > Hello Everybody. > > Before anything I want to say big thanks to the developers of OpenBSD, > for maintaining it. After some ~10 years of being the loyal OpenBSD > user, I never had any problem with OpenBSD itself, besides may be 2 or > three times. > It is impressive. Every other system I use gives problems from time to > time, so I am thanking you, guys, every time I type a command. > > Now onto the bitter part. For some reason, since, may be, AFAIR 5.2 > times, I do not see any browser that is working flawlessly under our > loved system. > Everything is happened on the same set of sites I use routinely everyday. > > I tried: > Firefox - bad, bad, bad. It fails 1000 times a day. On your machine, firefox couldn't be restarted 1000 times a day. (ok, not sure where my sense if irony is today...) ... > dmesg follows: > OpenBSD 5.6-current (GENERIC.MP) #340: Fri Aug 22 15:06:09 MDT 2014 > dera...@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC.MP > real mem = 1568260096 (1495MB) > avail mem = 1517772800 (1447MB) ... > cpu0 at mainbus0: apid 0 (boot processor) > cpu0: Intel(R) Celeron(R) CPU G530 @ 2.40GHz, 2394.94 MHz ok, how do I put this nicely... To run a modern browser, you need a modern computer. 1.5GB RAM and a celeron processor doesn't cut it. NOW, that doesn't cause CRASHES, but when you fix the crashes by cranking up your login.conf specs, you will be so far into swap you will wish your browser crashed. Modern browsers leak memory like everyone has 16GB and a quad-core proc, AND restarts their browser several times a day. Look at those same browsers on Windows (their target market), you see the same thing. The difference is, OpenBSD kicks out programs that exceed predefined limits, that's what you are most likely seeing. But most likely, login.conf will fix your crash problem, as I use firefox, Chromium and Thunderbird on my amd64 system (three-core, 4G RAM), and usually get a week or two uptime between shutdowns (because of hitting RAM limits). Nick.
[patch] www/faq/faq6.html: add otus(4), rsu(4), urtwn(4) to wireless networking list
Just noticed that these drivers are not listed Index: faq6.html === RCS file: /cvs/www/faq/faq6.html,v retrieving revision 1.318 diff -u -r1.318 faq6.html --- faq6.html7 Aug 2014 01:51:34 -1.318 +++ faq6.html23 Aug 2014 14:20:42 - @@ -2053,6 +2053,8 @@ Intel WiFi Link 4965/5100/5300 802.11a/b/g/Draft-N wireless. href="http://www.openbsd.org/cgi-bin/man.cgi?query=malo&sektion=4";>malo(4) Marvell Libertas 802.11b/g +href="http://www.openbsd.org/cgi-bin/man.cgi?query=otus&sektion=4";>otus(4) +Atheros USB 802.11a/g/n href="http://www.openbsd.org/cgi-bin/man.cgi?query=pgt&sektion=4";>pgt(4) Conexant/Intersil Prism GT Full-MAC 802.11a/b/g href="http://www.openbsd.org/cgi-bin/man.cgi?query=ral&sektion=4";>ral(4) @@ -2060,6 +2062,8 @@ Ralink Technology RT25x0 802.11a/b/g. (AP) href="http://www.openbsd.org/cgi-bin/man.cgi?query=ray&sektion=4";>ray(4) Raytheon Raylink/WebGear Aviator 802.11FH +href="http://www.openbsd.org/cgi-bin/man.cgi?query=rsu&sektion=4";>rsu(4) +Realtek RTL8188SU/RTL8192SU USB 802.11b/g/n href="http://www.openbsd.org/cgi-bin/man.cgi?query=rtw&sektion=4";>rtw(4) Realtek 8180 802.11b. (AP) href="http://www.openbsd.org/cgi-bin/man.cgi?query=rum&sektion=4";>rum(4) @@ -2072,6 +2076,8 @@ Conexant/Intersil PrismGT SoftMAC USB 802.11b/g href="http://www.openbsd.org/cgi-bin/man.cgi?query=urtw&sektion=4";>urtw(4) Realtek RTL8187L USB 802.11b/g +href="http://www.openbsd.org/cgi-bin/man.cgi?query=urtwn&sektion=4";>urtwn(4) +Realtek RTL8188CU/RTL8192CU USB 802.11b/g/n href="http://www.openbsd.org/cgi-bin/man.cgi?query=wi&sektion=4";>wi(4) Prism2/2.5/3. (AP) href="http://www.openbsd.org/cgi-bin/man.cgi?query=wpi&sektion=4";>wpi(4)
The rant about browsers
Hello Everybody. Before anything I want to say big thanks to the developers of OpenBSD, for maintaining it. After some ~10 years of being the loyal OpenBSD user, I never had any problem with OpenBSD itself, besides may be 2 or three times. It is impressive. Every other system I use gives problems from time to time, so I am thanking you, guys, every time I type a command. Now onto the bitter part. For some reason, since, may be, AFAIR 5.2 times, I do not see any browser that is working flawlessly under our loved system. Everything is happened on the same set of sites I use routinely everyday. I tried: Firefox - bad, bad, bad. It fails 1000 times a day. Chromium - it is better, in terms. Yes, it will not fail on the plain place (it is a Russian idiom, which means 'from nothing' or 'from no reason one can observe'), but left for some time it starts to be so slow... was forced to stay away from it too. but after all it is the only browser under OpenBSD that have a working lastpass plugin. (and I need lastpass, if I want to share my passwords between home and job computers) Seamonkey - potentially good project. but suffers from the same problems like firefox. although it is fails much much less, the frequency is still unacceptable for me. I also used xombrero and it was good, but again, from somewhere between 5.2 - 5.3 times it has started to fail with an unacceptable frequency. I know, I should write to upstream mailing lists of the projects I've mentioned above, but before that, I want to know if somebody else is suffering such problems and I am still sure maintatiners of the corresponding ports will do it better than me if they find it is a problem. -- With best regards, Gregory Edigarov dmesg follows: OpenBSD 5.6-current (GENERIC.MP) #340: Fri Aug 22 15:06:09 MDT 2014 dera...@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC.MP real mem = 1568260096 (1495MB) avail mem = 1517772800 (1447MB) mpath0 at root scsibus0 at mpath0: 256 targets mainbus0 at root bios0 at mainbus0: SMBIOS rev. 2.6 @ 0xeb170 (91 entries) bios0: vendor American Megatrends Inc. version "0701" date 07/04/2012 bios0: ASUSTeK COMPUTER INC. P8H61-M2 USB3 acpi0 at bios0: rev 2 acpi0: sleep states S0 S1 S3 S4 S5 acpi0: tables DSDT FACP APIC SSDT MCFG HPET acpi0: wakeup devices PS2K(S4) PS2M(S4) BR20(S3) EUSB(S4) USBE(S4) PEX0(S4) PEX1(S4) PEX3(S4) PEX5(S4) PEX6(S4) PEX7(S4) P0P1(S4) P0P2(S4) P0P3(S4) P0P4(S4) acpitimer0 at acpi0: 3579545 Hz, 24 bits acpimadt0 at acpi0 addr 0xfee0: PC-AT compat cpu0 at mainbus0: apid 0 (boot processor) cpu0: Intel(R) Celeron(R) CPU G530 @ 2.40GHz, 2394.94 MHz cpu0: FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,POPCNT,DEADLINE,XSAVE,NXE,LONG,LAHF,PERF,ITSC cpu0: 256KB 64b/line 8-way L2 cache cpu0: smt 0, core 0, package 0 mtrr: Pentium Pro MTRR support, 10 var ranges, 88 fixed ranges cpu0: apic clock running at 99MHz cpu0: mwait min=64, max=64, C-substates=0.2.1.1.0, IBE cpu1 at mainbus0: apid 2 (application processor) cpu1: Intel(R) Celeron(R) CPU G530 @ 2.40GHz, 2394.57 MHz cpu1: FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,POPCNT,DEADLINE,XSAVE,NXE,LONG,LAHF,PERF,ITSC cpu1: 256KB 64b/line 8-way L2 cache cpu1: smt 0, core 1, package 0 ioapic0 at mainbus0: apid 0 pa 0xfec0, version 20, 24 pins acpimcfg0 at acpi0 addr 0xe000, bus 0-63 acpihpet0 at acpi0: 14318179 Hz acpiprt0 at acpi0: bus 0 (PCI0) acpiprt1 at acpi0: bus 2 (PEX0) acpiprt2 at acpi0: bus 3 (PEX1) acpiprt3 at acpi0: bus 5 (PEX3) acpiprt4 at acpi0: bus 6 (PEX5) acpiprt5 at acpi0: bus -1 (PEX6) acpiprt6 at acpi0: bus -1 (PEX7) acpiprt7 at acpi0: bus 1 (P0P1) acpiprt8 at acpi0: bus -1 (P0P2) acpiprt9 at acpi0: bus -1 (P0P3) acpiprt10 at acpi0: bus -1 (P0P4) acpicpu0 at acpi0: C3, C2, C1, PSS acpicpu1 at acpi0: C3, C2, C1, PSS acpibtn0 at acpi0: PWRB acpivideo0 at acpi0: GFX0 acpivout0 at acpivideo0: DD02 cpu0: Enhanced SpeedStep 2394 MHz: speeds: 2400, 2300, 2200, 2100, 2000, 1900, 1800, 1700, 1600 MHz pci0 at mainbus0 bus 0 pchb0 at pci0 dev 0 function 0 "Intel Core 2G Host" rev 0x09 ppb0 at pci0 dev 1 function 0 "Intel Core 2G PCIE" rev 0x09: msi pci1 at ppb0 bus 1 vga1 at pci0 dev 2 function 0 "Intel HD Graphics 2000" rev 0x09 intagp at vga1 not configured inteldrm0 at vga1 drm0 at inteldrm0 drm: Memory usable by graphics device = 2048M inteldrm0: 1280x1024 wsdisplay0 at vga1 mux 1: console (std, vt100 emulation) wsdisplay0: screen 1-5 added (std, vt100 emulation) "Intel 6 Series MEI" rev 0x04 at pci0 dev 22 function 0 not configured ehci0 at pci0 dev 26 function 0 "Intel 6 Series USB" rev 0x05: apic 0 int 23 usb0 at ehci0: USB revision 2.0 uhub0 at usb0 "Intel EHCI
Re: OpenBSD 5.5-STABLE: Full Disk Encryption (bioctl) and Smard Cards
On August 23, 2014 4:33:55 AM CEST, "Артур Истомин" wrote: >On Fri, Aug 22, 2014 at 04:03:59PM -0700, Zach Leslie wrote: >> > However, I don't know how it is seen by the system and if it would >> > show up as a drive. Anyone in here is using a smart card to decrypt >> > volumes at boot? >> >> You could use a YubiKey with a static long password to unlock the >boot >> volume. > >[offtop] > >Are there any YubiKey-like devices that can contain many static >password, not one like YubiKey? Not sure it helps, but mine contains two... /Alexander
Re: iked troubles, SA not installed
On 2014-08-21, Vincent Gross wrote: > here is the routing table on the gateway once S[AP] are installed: > > Encap: > Source Port DestinationPort Proto > SA(Address/Proto/Type/Direction) > 192.168.55.220/32 0 192.168.56.1/320 0 > 37.160.166.168/esp/use/in > 192.168.56.1/320 192.168.55.220/32 0 0 > 37.160.166.168/esp/require/out > default0 default0 0 none/esp/deny/out > > Yet, tcpdump on gateway's enc0 shows this: > > tcpdump: listening on enc0, link-type ENC > tcpdump: WARNING: compensating for unaligned libpcap packets > 11:29:00.455369 (authentic,confidential): SPI 0xa5ba5ce9: 79.143.250.153.22 > > 37.160.166.168.16215: P 1027357934:1027357978(44) ack 3953089614 win 2112 (DF) > [tos 0x10] (encap) > 11:29:00.456355 (authentic,confidential): SPI 0xa5ba5ce9: 79.143.250.153.22 > > 37.160.166.168.16215: P 44:88(44) ack 1 win 2112 (DF) [tos 0x10] (encap) I've reported problems like this before, where traffic is handled by IPsec that shouldn't be - and mostly (or possibly always) connected with IPsec flows that restrict traffic by protocol. > When I got this dump, I already had an SSH connection between laptop and > gateway, and I tried to connect to gateway's 222/tcp using telnet. > > In my previous message, I put a tcpdump trace showing what happens when > I try to establish a TCP connection: I had the TCP handshake completed > over raw IP, the laptop sent its first data packet, but I had no > response whatsoever, just a bunch of ESP packets. > > So This is what I conclude form all that stuff: > 1) IPSec parameters are negociated between ikeds > 2) gateway installs SPs and SAs > 3) TCP handshake goes on raw IP, no problem > 4) gateway routes all established TCP flows through IPSec, including those > already established and not matched by the installed SPs ... > > I ran a test over UDP using inetd echo on gateway, and nc -u on the > laptop. After the gateway installed the SAs and SPs, I had no problem > having the data I sent form the laptop to the gateway echoed back, so > whatever is going on during the routing phase, it leaves UDP traffic > alone. I have seen it with UDP as well, at least DNS and NTP traffic. > I will update both systems tonight with the latest snapshot, and seen if > the problem persists. It has persisted for at least several years :(
Re: Cloning an OpenBSD system (and potential FAQ (4.15) error?)
On Sat, Aug 23, 2014 at 6:21 AM, Stuart Henderson wrote: > It may be easier to installboot(8) after copying. Yeah I used installboot -- "Don't eat anything you've ever seen advertised on TV" - Michael Pollan, author of "In Defense of Food"
Re: New queueing system and HZ value limits
On 2014-08-22, Henning Brauer wrote: > * Stuart Henderson [2014-08-22 13:51]: >> On 2014-08-22, Henning Brauer wrote: >> > * Federico Giannici [2014-08-22 09:51]: >> >> On 08/22/14 08:22, Henning Brauer wrote: >> >> >* Adam Thompson [2014-08-21 19:13]: >> >> >>Unless I've mis-understood all the emails and reports about this, it >> >> >>affects low-bandwidth queues, not low-bandwidth interfaces. >> >> >>In other words, limiting traffic to 50Mbps on a 1Gb link will work >> >> >>fine, limiting it to 50kbps on the same link will not. >> >> >>Yes/no? >> >> >pretty much. >> >> I can imagine that it could be rather complicated to give the exact >> >> numbers, >> >> but can you give me an idea where the problem comes from, and maybe where >> >> I >> >> can find more info about it? >> > kinda obvious: BW measurement and go/holdoff decision is (at most) once per >> > tick. ticks @ HZ, aka 100 ticks per second with HZ=100. If the NIC can >> > transfer "too much" data within one tick, the bw shaping becomes >> > inaccurate. Obviously worse the bigger the difference between >> > interface speed and desired queue speed is. >> Any idea why this was so much less of a problem with altq? > > it wasn't... the hfsc core was the same, and cbq worked exactly the same > way too. > > People might not have paid as much attention? I dunno. If anything I'd expect altq to be less accurate as IIRC it used getmicrouptime rather than microuptime But somehow, my setup with 512K-1Mb queues (pppoe with pppoedev on em0, 100Mb link on a 1Gb nic) worked ok with altq with default HZ.
Re: Cloning an OpenBSD system (and potential FAQ (4.15) error?)
On 2014-08-22, Maurice McCarthy wrote: > Hi, > > /boot is found by block number and offset of its inode so I think the root > partition should be copied using dd. It may be easier to installboot(8) after copying.
Re: OpenBSD 5.5-STABLE: Full Disk Encryption (bioctl) and Smard Cards
On 2014-08-22, Julien Meister wrote: > Thank you very much. > > So there is really really no way for the system to retrieve the key stored > on the smart card (using GnuPG) at boot in order to decrypt > the volumes? Correct, you can't run application programs like GnuPG before the system has booted.
Re: amavisd uses high cpu usage?
On 2014/08/23 09:02, Indunil Jayasooriya wrote: > > > Hi Stuart, > > > > amavisd-new runs fine for me on OpenBSD without particularly high > CPU use. > > > I am very glad to hear that it is running fine on my favourite > Operating system OpenBSD. > > is Amavisd-new running on OpenBSD 5.5 ? I've had it running on every version since about 5.2. > > I did a debug with the command " /usr/local/sbin/amavisd debug " > (I set $log_level = 5 in /etc/amavisd.conf file) > > it says > > Segmentation fault > > Then, I uncommented @bypass_spam_checks_maps = (1); in /etc/ > amavisd.conf file. > > Pls see below > > > # @bypass_virus_checks_maps = (1); # controls running of anti-virus > code > @bypass_spam_checks_maps = (1); # controls running of anti-spam code > # $bypass_decode_parts = 1; # controls running of decoders& > dearchivers > > > Then. restarted amavisd ( /etc/rc.d/amavisd restart ) . Then, It > started working.. > I did a debug with the command " /usr/local/sbin/amavisd debug " again > > then, it gave this. > > The amavisd daemon is already running, PID: [4909] "amavisd debug" runs a standalone copy, displaying log entries on screen rather than to a file. You should run it when amavisd is not already running. > I think may be something is wrong with perl modules. Could be. How did you install things? Is this a fresh 5.5 install or an upgrade from an earlier version? Did you upgrade all packages?
