Re: MS DNS doesn't answer to CURRENT under Hyper-V
Hi! r285785 still isn't MFCed. RC2 is coming soon. 2015-07-23 10:54 GMT+03:00 Pavel Timofeev tim...@gmail.com: Ok, sorry! 2015-07-23 7:51 GMT+03:00 Wei Hu w...@microsoft.com: The TCP offloading is still working on these platforms. There is no flag to distinguish UDP and TCP offloading, so the RXCSUM and TXCSUM are still set. Let me know if there is any other way to show it properly. Thanks, Wei -Original Message- From: Pavel Timofeev [mailto:tim...@gmail.com] Sent: Wednesday, July 22, 2015 9:04 PM To: Wei Hu w...@microsoft.com Cc: Slawa Olhovchenkov s...@zxy.spb.ru; freebsd-curr...@freebsd.org; freebsd-virtualization@freebsd.org Subject: Re: MS DNS doesn't answer to CURRENT under Hyper-V Hi! I see you have done the code for disabling UDP checksum offloading when running on the Hyper-V on Windows Server 2012 and earlier hosts https://svnweb.freebsd.org/base?view=revisionrevision=285785 I tried new CURRENT and it works. Thank you! A small note here: while it disables and works it still shows RXCSUM and TSCSUM in iface's options: root@proxy:/usr/src # ifconfig hn0 hn0: flags=8843UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST metric 0 mtu 1500 options=31bRXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,TSO4,TSO6 ether 00:15:5d:02:9c:09 inet 192.168.25.26 netmask 0xffc0 broadcast 192.168.25.63 nd6 options=29PERFORMNUD,IFDISABLED,AUTO_LINKLOCAL Is it possible to hide it automatically if it's disabled by new code? 2015-07-13 11:06 GMT+03:00 Wei Hu w...@microsoft.com: We have root caused the problem. This issue happens on the Hyper-Vs on Windows Server 2012 (Win 8.0) and earlier releases. On these releases, the UPD checksum offloading on host side does not work properly. The workaround is to disable UPD checksum offloading in the FreeBSD guest through 'ifconfig'. We are also working on a patch to turn off UPD checksum offloading in the netvsc driver when detecting the Hyper-V releases. The UDP checksum offloading works fine on Windows Server 2012R2 and Win 8.1 hosts. Thanks Pavel and Slawa for the support. Wei -Original Message- From: owner-freebsd-virtualizat...@freebsd.org [mailto:owner-freebsd- virtualizat...@freebsd.org] On Behalf Of Pavel Timofeev Sent: Wednesday, July 8, 2015 4:06 PM To: Slawa Olhovchenkov Cc: freebsd-curr...@freebsd.org; freebsd-virtualization@freebsd.org Subject: Re: MS DNS doesn't answer to CURRENT under Hyper-V Ok, r284746 is the root of the problem. MS DNS works under r284745 and doesn't work under r284746. Slawa, what should I look at in wireshark output? 2015-07-07 18:49 GMT+03:00 Slawa Olhovchenkov s...@zxy.spb.ru: On Tue, Jul 07, 2015 at 06:04:46PM +0300, Pavel Timofeev wrote: Well, turning off checksum offloading by `ifconfig hn0 -txcsum -rxcsum` definitely helps. As for tcpdump I'm not completely sure if I did it right, but I see bad udp cksum phrase: # tcpdump -i hn0 -vvv -nn udp dst port 53 tcpdump: listening on hn0, link-type EN10MB (Ethernet), capture size 262144 bytes 18:01:19.139994 IP (tos 0x0, ttl 64, id 61218, offset 0, flags [none], proto UDP (17), length 51) 192.168.25.26.45683 192.168.25.3.53: [bad udp cksum 0xb39e - 0xf210!] 52886+ A? ya.ru. (23) 18:01:24.140544 IP (tos 0x0, ttl 64, id 17293, offset 0, flags [none], proto UDP (17), length 51) 192.168.25.26.12575 192.168.25.3.53: [bad udp cksum 0xb39e - 0x7365!] 52886+ A? ya.ru. (23) tcpdump bad udp cksum is normal on FreeBSD host in case checksum offload (and may be need only for help finding issuse in code). Need wireshark capturing from MS DNS host (or from mirroring port). ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization- unsubscr...@freebsd.org ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
Re: MS DNS doesn't answer to CURRENT under Hyper-V
Ok, sorry! 2015-07-23 7:51 GMT+03:00 Wei Hu w...@microsoft.com: The TCP offloading is still working on these platforms. There is no flag to distinguish UDP and TCP offloading, so the RXCSUM and TXCSUM are still set. Let me know if there is any other way to show it properly. Thanks, Wei -Original Message- From: Pavel Timofeev [mailto:tim...@gmail.com] Sent: Wednesday, July 22, 2015 9:04 PM To: Wei Hu w...@microsoft.com Cc: Slawa Olhovchenkov s...@zxy.spb.ru; freebsd-curr...@freebsd.org; freebsd-virtualization@freebsd.org Subject: Re: MS DNS doesn't answer to CURRENT under Hyper-V Hi! I see you have done the code for disabling UDP checksum offloading when running on the Hyper-V on Windows Server 2012 and earlier hosts https://svnweb.freebsd.org/base?view=revisionrevision=285785 I tried new CURRENT and it works. Thank you! A small note here: while it disables and works it still shows RXCSUM and TSCSUM in iface's options: root@proxy:/usr/src # ifconfig hn0 hn0: flags=8843UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST metric 0 mtu 1500 options=31bRXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,TSO4,TSO6 ether 00:15:5d:02:9c:09 inet 192.168.25.26 netmask 0xffc0 broadcast 192.168.25.63 nd6 options=29PERFORMNUD,IFDISABLED,AUTO_LINKLOCAL Is it possible to hide it automatically if it's disabled by new code? 2015-07-13 11:06 GMT+03:00 Wei Hu w...@microsoft.com: We have root caused the problem. This issue happens on the Hyper-Vs on Windows Server 2012 (Win 8.0) and earlier releases. On these releases, the UPD checksum offloading on host side does not work properly. The workaround is to disable UPD checksum offloading in the FreeBSD guest through 'ifconfig'. We are also working on a patch to turn off UPD checksum offloading in the netvsc driver when detecting the Hyper-V releases. The UDP checksum offloading works fine on Windows Server 2012R2 and Win 8.1 hosts. Thanks Pavel and Slawa for the support. Wei -Original Message- From: owner-freebsd-virtualizat...@freebsd.org [mailto:owner-freebsd- virtualizat...@freebsd.org] On Behalf Of Pavel Timofeev Sent: Wednesday, July 8, 2015 4:06 PM To: Slawa Olhovchenkov Cc: freebsd-curr...@freebsd.org; freebsd-virtualization@freebsd.org Subject: Re: MS DNS doesn't answer to CURRENT under Hyper-V Ok, r284746 is the root of the problem. MS DNS works under r284745 and doesn't work under r284746. Slawa, what should I look at in wireshark output? 2015-07-07 18:49 GMT+03:00 Slawa Olhovchenkov s...@zxy.spb.ru: On Tue, Jul 07, 2015 at 06:04:46PM +0300, Pavel Timofeev wrote: Well, turning off checksum offloading by `ifconfig hn0 -txcsum -rxcsum` definitely helps. As for tcpdump I'm not completely sure if I did it right, but I see bad udp cksum phrase: # tcpdump -i hn0 -vvv -nn udp dst port 53 tcpdump: listening on hn0, link-type EN10MB (Ethernet), capture size 262144 bytes 18:01:19.139994 IP (tos 0x0, ttl 64, id 61218, offset 0, flags [none], proto UDP (17), length 51) 192.168.25.26.45683 192.168.25.3.53: [bad udp cksum 0xb39e - 0xf210!] 52886+ A? ya.ru. (23) 18:01:24.140544 IP (tos 0x0, ttl 64, id 17293, offset 0, flags [none], proto UDP (17), length 51) 192.168.25.26.12575 192.168.25.3.53: [bad udp cksum 0xb39e - 0x7365!] 52886+ A? ya.ru. (23) tcpdump bad udp cksum is normal on FreeBSD host in case checksum offload (and may be need only for help finding issuse in code). Need wireshark capturing from MS DNS host (or from mirroring port). ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization- unsubscr...@freebsd.org ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
Re: MS DNS doesn't answer to CURRENT under Hyper-V
Hi! I see you have done the code for disabling UDP checksum offloading when running on the Hyper-V on Windows Server 2012 and earlier hosts https://svnweb.freebsd.org/base?view=revisionrevision=285785 I tried new CURRENT and it works. Thank you! A small note here: while it disables and works it still shows RXCSUM and TSCSUM in iface's options: root@proxy:/usr/src # ifconfig hn0 hn0: flags=8843UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST metric 0 mtu 1500 options=31bRXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,TSO4,TSO6 ether 00:15:5d:02:9c:09 inet 192.168.25.26 netmask 0xffc0 broadcast 192.168.25.63 nd6 options=29PERFORMNUD,IFDISABLED,AUTO_LINKLOCAL Is it possible to hide it automatically if it's disabled by new code? 2015-07-13 11:06 GMT+03:00 Wei Hu w...@microsoft.com: We have root caused the problem. This issue happens on the Hyper-Vs on Windows Server 2012 (Win 8.0) and earlier releases. On these releases, the UPD checksum offloading on host side does not work properly. The workaround is to disable UPD checksum offloading in the FreeBSD guest through 'ifconfig'. We are also working on a patch to turn off UPD checksum offloading in the netvsc driver when detecting the Hyper-V releases. The UDP checksum offloading works fine on Windows Server 2012R2 and Win 8.1 hosts. Thanks Pavel and Slawa for the support. Wei -Original Message- From: owner-freebsd-virtualizat...@freebsd.org [mailto:owner-freebsd- virtualizat...@freebsd.org] On Behalf Of Pavel Timofeev Sent: Wednesday, July 8, 2015 4:06 PM To: Slawa Olhovchenkov Cc: freebsd-curr...@freebsd.org; freebsd-virtualization@freebsd.org Subject: Re: MS DNS doesn't answer to CURRENT under Hyper-V Ok, r284746 is the root of the problem. MS DNS works under r284745 and doesn't work under r284746. Slawa, what should I look at in wireshark output? 2015-07-07 18:49 GMT+03:00 Slawa Olhovchenkov s...@zxy.spb.ru: On Tue, Jul 07, 2015 at 06:04:46PM +0300, Pavel Timofeev wrote: Well, turning off checksum offloading by `ifconfig hn0 -txcsum -rxcsum` definitely helps. As for tcpdump I'm not completely sure if I did it right, but I see bad udp cksum phrase: # tcpdump -i hn0 -vvv -nn udp dst port 53 tcpdump: listening on hn0, link-type EN10MB (Ethernet), capture size 262144 bytes 18:01:19.139994 IP (tos 0x0, ttl 64, id 61218, offset 0, flags [none], proto UDP (17), length 51) 192.168.25.26.45683 192.168.25.3.53: [bad udp cksum 0xb39e - 0xf210!] 52886+ A? ya.ru. (23) 18:01:24.140544 IP (tos 0x0, ttl 64, id 17293, offset 0, flags [none], proto UDP (17), length 51) 192.168.25.26.12575 192.168.25.3.53: [bad udp cksum 0xb39e - 0x7365!] 52886+ A? ya.ru. (23) tcpdump bad udp cksum is normal on FreeBSD host in case checksum offload (and may be need only for help finding issuse in code). Need wireshark capturing from MS DNS host (or from mirroring port). ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization- unsubscr...@freebsd.org ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
Re: MS DNS doesn't answer to CURRENT under Hyper-V
Ok, r284746 is the root of the problem. MS DNS works under r284745 and doesn't work under r284746. Slawa, what should I look at in wireshark output? 2015-07-07 18:49 GMT+03:00 Slawa Olhovchenkov s...@zxy.spb.ru: On Tue, Jul 07, 2015 at 06:04:46PM +0300, Pavel Timofeev wrote: Well, turning off checksum offloading by `ifconfig hn0 -txcsum -rxcsum` definitely helps. As for tcpdump I'm not completely sure if I did it right, but I see bad udp cksum phrase: # tcpdump -i hn0 -vvv -nn udp dst port 53 tcpdump: listening on hn0, link-type EN10MB (Ethernet), capture size 262144 bytes 18:01:19.139994 IP (tos 0x0, ttl 64, id 61218, offset 0, flags [none], proto UDP (17), length 51) 192.168.25.26.45683 192.168.25.3.53: [bad udp cksum 0xb39e - 0xf210!] 52886+ A? ya.ru. (23) 18:01:24.140544 IP (tos 0x0, ttl 64, id 17293, offset 0, flags [none], proto UDP (17), length 51) 192.168.25.26.12575 192.168.25.3.53: [bad udp cksum 0xb39e - 0x7365!] 52886+ A? ya.ru. (23) tcpdump bad udp cksum is normal on FreeBSD host in case checksum offload (and may be need only for help finding issuse in code). Need wireshark capturing from MS DNS host (or from mirroring port). ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
Re: MS DNS doesn't answer to CURRENT under Hyper-V
No, I don't even know what VMDepot is, sorry! Didn't know ;) 2015-07-07 16:58 GMT+03:00 Glen Barber g...@freebsd.org: BTW, are you using the images from VMDepot? If so, this was MFC'd after the recent images. I can easily regenerate an updated image, if you are using those. Glen On Tue, Jul 07, 2015 at 04:55:39PM +0300, Pavel Timofeev wrote: Wow, r284746 was MFCed to 10 STABLE! I should hurry up! 2015-07-07 16:26 GMT+03:00 Pavel Timofeev tim...@gmail.com: Ok, I'll try r284745 and then r284746 and see what happens. 2015-07-07 16:06 GMT+03:00 Wei Hu w...@microsoft.com: -Original Message- From: owner-freebsd-virtualizat...@freebsd.org [mailto:owner-freebsd- virtualizat...@freebsd.org] On Behalf Of Pavel Timofeev Sent: Tuesday, July 7, 2015 7:51 PM To: freebsd-curr...@freebsd.org; freebsd-virtualization@freebsd.org Subject: MS DNS doesn't answer to CURRENT under Hyper-V Hi! I have a test virtual machine which runs CURRENT under Hyper-V. It's amd64 r285198 now. It can't get any response from MS DNS server. Well, it could two or three weeks ago, but after upgrade it's not able to do it anymore. Google DNS answers without problems meanwhile (sic!). What I do: # host google.ru 192.168.25.3 I see that MS DNS (192.168.25.3) server receives these packets, but ignores them. And no matter how my system asks MS DNS. Every daemon can't get response too. I know that nothing was changed in MS DNS server. No doubt. Then I tried different available CURRENT snapshot ISOs. FreeBSD-11.0-CURRENT-amd64-20150630-r284969-disc1.iso - MS DNS does not answer. FreeBSD-11.0-CURRENT-amd64-20150625-r284814-disc1.iso - MS DNS does not answer. FreeBSD-11.0-CURRENT-amd64-20150618-r284544-disc1.iso - MS DNS answers! So something was committed to CURRENT between 20150618 and 20150625. This something ruins communication with MS DNS. There was a commit for Hyper-V TSO and checksum offloading support (r284746) on June 24th. I think this commit is the cause. Can you verify the MS DNS behavior between The builds of June 23rd and 24th? I will take a look of this issue tomorrow. Thanks, Wei ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
Re: MS DNS doesn't answer to CURRENT under Hyper-V
Well, turning off checksum offloading by `ifconfig hn0 -txcsum -rxcsum` definitely helps. As for tcpdump I'm not completely sure if I did it right, but I see bad udp cksum phrase: # tcpdump -i hn0 -vvv -nn udp dst port 53 tcpdump: listening on hn0, link-type EN10MB (Ethernet), capture size 262144 bytes 18:01:19.139994 IP (tos 0x0, ttl 64, id 61218, offset 0, flags [none], proto UDP (17), length 51) 192.168.25.26.45683 192.168.25.3.53: [bad udp cksum 0xb39e - 0xf210!] 52886+ A? ya.ru. (23) 18:01:24.140544 IP (tos 0x0, ttl 64, id 17293, offset 0, flags [none], proto UDP (17), length 51) 192.168.25.26.12575 192.168.25.3.53: [bad udp cksum 0xb39e - 0x7365!] 52886+ A? ya.ru. (23) 2015-07-07 16:09 GMT+03:00 Slawa Olhovchenkov s...@zxy.spb.ru: On Tue, Jul 07, 2015 at 02:50:49PM +0300, Pavel Timofeev wrote: Hi! I have a test virtual machine which runs CURRENT under Hyper-V. It's amd64 r285198 now. It can't get any response from MS DNS server. Well, it could two or three weeks ago, but after upgrade it's not able to do it anymore. Google DNS answers without problems meanwhile (sic!). What I do: # host google.ru 192.168.25.3 I see that MS DNS (192.168.25.3) server receives these packets, but ignores them. And no matter how my system asks MS DNS. Every daemon can't get response too. I know that nothing was changed in MS DNS server. No doubt. Then I tried different available CURRENT snapshot ISOs. FreeBSD-11.0-CURRENT-amd64-20150630-r284969-disc1.iso - MS DNS does not answer. FreeBSD-11.0-CURRENT-amd64-20150625-r284814-disc1.iso - MS DNS does not answer. FreeBSD-11.0-CURRENT-amd64-20150618-r284544-disc1.iso - MS DNS answers! So something was committed to CURRENT between 20150618 and 20150625. This something ruins communication with MS DNS. Then I tried latest FreeBSD-11.0-CURRENT-amd64-20150630-r284969-disc1.iso on bare metal - MS DNS answered! Looks like that something is related to Hyper-V code. Maybe it changes packets somehow? I can gather and provide more info (tcpdump?) if you ask, it's not a problem! Author: whu Date: Wed Jun 24 06:01:29 2015 New Revision: 284746 URL: https://svnweb.freebsd.org/changeset/base/284746 Log: TSO and checksum offloading support for Netvsc driver on Hyper-V. = Try tcpdump/wireshark on FreeBSD and MS DNS host. Check validating IP/UDP checksums. Try off checksum offloading on network interface (ifconfig ifname -txcsum -rxcsum) ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
vt(4) performance on Hyper-V
Hi! Just in case, just so you know. I've tried FreeBSD 10-STABLE r270125 with vt(4) on Hyper-V (Win Server 2012 DataCenter). Its performance is extremely bad (while syscons is ok). Here is video http://www.youtube.com/watch?v=62CIKCJ7bog P.S. vt(4) is a bit faster on VirtualBox though when I tried it last time. ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
Re: vt(4) performance on Hyper-V
Wow, thank you! It helped. Sorry for the noise, it's even written in man vt. 2014-08-18 12:25 GMT+04:00 Luigi Rizzo ri...@iet.unipi.it: On Mon, Aug 18, 2014 at 9:08 AM, Pavel Timofeev tim...@gmail.com wrote: Hi! Just in case, just so you know. I've tried FreeBSD 10-STABLE r270125 with vt(4) on Hyper-V (Win Server 2012 DataCenter). Its performance is extremely bad (while syscons is ok). Here is video http://www.youtube.com/watch?v=62CIKCJ7bog i think you need to put the following in /boot/loader.conf hw.vga.textmode=1 otherwise it is very slow even on bare metal. cheers luigi P.S. vt(4) is a bit faster on VirtualBox though when I tried it last time. ___ freebsd-sta...@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to freebsd-stable-unsubscr...@freebsd.org -- -+--- Prof. Luigi RIZZO, ri...@iet.unipi.it . Dip. di Ing. dell'Informazione http://www.iet.unipi.it/~luigi/. Universita` di Pisa TEL +39-050-2211611 . via Diotisalvi 2 Mobile +39-338-6809875 . 56122 PISA (Italy) -+--- ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
Re: Running `sysctl -a` causes crash (Hyper-V)
I hope all these changes you have done will be included in FreeBSD 10-RELEASE. ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
Re: Hyper-V
I figured out another operations that cause caclru message appearance, such as 'top -n', car /proc/0/status, etc. But I tried these on 9.2-RELEASE LIVE CD and I saw calcru messages. So it seems like it's not related to brand new hyper-v integrated components. It's related to whole FreeBSD system. Which part? I don't know. 2013/11/12 Pavel Timofeev tim...@gmail.com: So, we catched the bug that caused crash. Thanks to all. Cool! As regards the messages about calcru: I tried to install CentOS 6.4 on new virt machine on same host system. Linux doesn't have any strange messages about time. I tried to migrate FreeBSD and CentOS virt machines to another host which has another hardware. Same story. For example, try to boot beta-3 in LIVE CD mode. You won't see calcru messages. But if you try to run 'dclient hn0' you'll see calcru message storm for some time. On already installed system I can cause messages by doing service zabbix_server start (even if it's not allowed in rc.conf (!)) But I can start mysql-server, zabbix_agentd, apache22 and it doesn't cause appearance of calcru messages. Any ideas? How and where can I dig? 2013/11/6 Pavel Timofeev tim...@gmail.com: Small fix: kern.timecounter.hardware=TSC covered messages, but now it looks stange % time sleep 30 20.24 real 0.00 user 0.00 sys 2013/11/6 Pavel Timofeev tim...@gmail.com: I already tried disabling Hyper-v time syncing, setting kern.timecounter.hardware to different type from kern.timecounter.choice, setting kern.hz to different values. But nothing helped. I'll try to install CentOS 6.4 on this machine and see what happens. 2013/11/6 Pavel Timofeev tim...@gmail.com: Does anybody know how to disable Hyper-v time syncing it guest FreeBSD OS? I'd like to try this one. 2013/11/6 Pavel Timofeev tim...@gmail.com: Thank you very much! I started to dig, and found that I can reboot my server running under Hyper-V by doing 'sysctl -a'. Can anybody confirm same behavoiur? FreeBSD 10.0-BETA3, Win 2012 Datacenter 2013/11/5 Julian Elischer jul...@freebsd.org: On 11/4/13, 11:13 PM, Pavel Timofeev wrote: Hi! I upgraded to 10.0-BETA3 but still see messages like Tue Nov 5 11:09:05 MSK 2013 calcru: runtime went backwards from 1728 usec to 889 usec for pid 981 (zabbix_server) calcru: runtime went backwards from 1229 usec to 703 usec for pid 976 (zabbix_server) calcru: runtime went backwards from 1019 usec to 595 usec for pid 978 (zabbix_server) calcru: runtime went backwards from 3041 usec to 1611 usec for pid 969 (zabbix_server) calcru: runtime went backwards from 1078 usec to 595 usec for pid 975 (zabbix_server) calcru: runtime went backwards from 1114 usec to 617 usec for pid 980 (zabbix_server) calcru: runtime went backwards from 1086 usec to 602 usec for pid 977 (zabbix_server) calcru: runtime went backwards from 1195 usec to 782 usec for pid 970 (zabbix_server) calcru: runtime went backwards from 3206 usec to 2157 usec for pid 979 (zabbix_server) calcru: runtime went backwards from 2179 usec to 1542 usec for pid 972 (zabbix_server) calcru: runtime went backwards from 1833 usec to 927 usec for pid 1026 (cron) calcru: runtime went backwards from 1091 usec to 551 usec for pid 1022 (sendmail) calcru: runtime went backwards from 823 usec to 416 usec for pid 1022 (sendmail) calcru: runtime went backwards from 4691 usec to 2735 usec for pid 1011 (sendmail) calcru: runtime went backwards from 1162 usec to 587 usec for pid 1011 (sendmail) calcru: runtime went backwards from 947 usec to 478 usec for pid 1010 (httpd) calcru: runtime went backwards from 895 usec to 452 usec for pid 1009 (httpd) calcru: runtime went backwards from 954 usec to 482 usec for pid 1008 (httpd) calcru: runtime went backwards from 909 usec to 459 usec for pid 1007 (httpd) calcru: runtime went backwards from 1214 usec to 653 usec for pid 1006 (httpd) calcru: runtime went backwards from 201793 usec to 102672 usec for pid 996 (httpd) calcru: runtime went backwards from 1105 usec to 1064 usec for pid 991 (sshd) calcru: runtime went backwards from 45196 usec to 23221 usec for pid 974 (zabbix_server) calcru: runtime went backwards from 2192 usec to 1436 usec for pid 973 (zabbix_server) calcru: runtime went backwards from 605 usec to 306 usec for pid 968 (zabbix_server) calcru: runtime went backwards from 626 usec to 316 usec for pid 967 (zabbix_server) calcru: runtime went backwards from 773 usec to 391 usec for pid 966 (zabbix_server) calcru: runtime went backwards from 663 usec to 335 usec for pid 965 (zabbix_server) calcru: runtime went backwards from 799 usec to 404 usec for pid 964 (zabbix_server) calcru: runtime went backwards from 37723 usec to 19422 usec for pid 963 (zabbix_server) calcru: runtime went backwards from 49425 usec to 26265 usec for pid 962 (zabbix_server) calcru: runtime went backwards from 38883 usec to 20993 usec for pid 961 (zabbix_server
Running `sysctl -a` causes crash (Hyper-V)
Hi! Can anybody confirm the subject? Every time when I run `sysctl -a` get this http://pastebin.com/Chf9qQTD Additional vmcore: http://yadi.sk/d/EAvZe3vZCVvp6 core.txt: http://yadi.sk/d/IBLJ8NVTCVvpn It's on Windows Server 2012 Datacenter ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
Re: Running `sysctl -a` causes crash (Hyper-V)
I'm sorry, could you explain how to do it? Which one is the 'core file', vmcore or core.txt? 2013/11/11 Sergey Kandaurov pluk...@gmail.com: On 11 November 2013 16:38, Pavel Timofeev tim...@gmail.com wrote: Hi! Can anybody confirm the subject? Every time when I run `sysctl -a` get this http://pastebin.com/Chf9qQTD Can you show 'p vm_guest' from core file? -- wbr, pluknet ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
Re: Running `sysctl -a` causes crash (Hyper-V)
root@bsdtest:~ # kgdb /boot/kernel/kernel ./vmcore.1 GNU gdb 6.1.1 [FreeBSD] Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type show copying to see the conditions. There is absolutely no warranty for GDB. Type show warranty for details. This GDB was configured as amd64-marcel-freebsd... Unread portion of the kernel message buffer: Fatal trap 12: page fault while in kernel mode cpuid = 1; apic id = 01 fault virtual address = 0x0 fault code = supervisor read data, page not present instruction pointer = 0x20:0x8095d79b stack pointer = 0x28:0xfe003d31c8f0 frame pointer = 0x28:0xfe003d31c900 code segment= base 0x0, limit 0xf, type 0x1b = DPL 0, pres 1, long 1, def32 0, gran 1 processor eflags= interrupt enabled, resume, IOPL = 0 current process = 777 (sysctl) trap number = 12 panic: page fault cpuid = 1 KDB: stack backtrace: #0 0x808e7a90 at kdb_backtrace+0x60 #1 0x808af575 at panic+0x155 #2 0x80c8e272 at trap_fatal+0x3a2 #3 0x80c8e549 at trap_pfault+0x2c9 #4 0x80c8dcd6 at trap+0x5e6 #5 0x80c74f72 at calltrap+0x8 #6 0x808ea529 at sysctl_kern_vm_guest+0x29 #7 0x808baacf at sysctl_root+0x24f #8 0x808bb088 at userland_sysctl+0x1d8 #9 0x808bae74 at sys___sysctl+0x74 #10 0x80c8eb67 at amd64_syscall+0x357 #11 0x80c7525b at Xfast_syscall+0xfb Uptime: 30s Dumping 89 out of 1001 MB:..18%..36%..54%..72%..89% Reading symbols from /boot/kernel/fuse.ko.symbols...done. Loaded symbols for /boot/kernel/fuse.ko.symbols Reading symbols from /boot/kernel/fdescfs.ko.symbols...done. Loaded symbols for /boot/kernel/fdescfs.ko.symbols #0 doadump (textdump=value optimized out) at pcpu.h:219 219 pcpu.h: No such file or directory. in pcpu.h (kgdb) p vm_guest $1 = 3 Current language: auto; currently minimal (kgdb) 2013/11/11 Sergey Kandaurov pluk...@gmail.com: [Please, don't top-post.] On 11 November 2013 17:06, Pavel Timofeev tim...@gmail.com wrote: I'm sorry, could you explain how to do it? Which one is the 'core file', vmcore or core.txt? It's vmcore. kgdb /path/to/kernel /path/to/vmcore 2013/11/11 Sergey Kandaurov pluk...@gmail.com: On 11 November 2013 16:38, Pavel Timofeev tim...@gmail.com wrote: Hi! Can anybody confirm the subject? Every time when I run `sysctl -a` get this http://pastebin.com/Chf9qQTD Can you show 'p vm_guest' from core file? -- wbr, pluknet ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
Re: Running `sysctl -a` causes crash (Hyper-V)
We'll see, I started kernel rebuilding. 2013/11/11 Sergey Kandaurov pluk...@gmail.com: On 11 November 2013 18:13, Pavel Timofeev tim...@gmail.com wrote: [...] (kgdb) p vm_guest $1 = 3 Current language: auto; currently minimal (kgdb) What if you try this change? Index: sys/kern/subr_param.c === --- sys/kern/subr_param.c(revision 257695) +++ sys/kern/subr_param.c(working copy) @@ -153,6 +153,7 @@ none, generic, xen, +hv, NULL }; ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
Re: Hyper-V
Thank you very much! I started to dig, and found that I can reboot my server running under Hyper-V by doing 'sysctl -a'. Can anybody confirm same behavoiur? FreeBSD 10.0-BETA3, Win 2012 Datacenter 2013/11/5 Julian Elischer jul...@freebsd.org: On 11/4/13, 11:13 PM, Pavel Timofeev wrote: Hi! I upgraded to 10.0-BETA3 but still see messages like Tue Nov 5 11:09:05 MSK 2013 calcru: runtime went backwards from 1728 usec to 889 usec for pid 981 (zabbix_server) calcru: runtime went backwards from 1229 usec to 703 usec for pid 976 (zabbix_server) calcru: runtime went backwards from 1019 usec to 595 usec for pid 978 (zabbix_server) calcru: runtime went backwards from 3041 usec to 1611 usec for pid 969 (zabbix_server) calcru: runtime went backwards from 1078 usec to 595 usec for pid 975 (zabbix_server) calcru: runtime went backwards from 1114 usec to 617 usec for pid 980 (zabbix_server) calcru: runtime went backwards from 1086 usec to 602 usec for pid 977 (zabbix_server) calcru: runtime went backwards from 1195 usec to 782 usec for pid 970 (zabbix_server) calcru: runtime went backwards from 3206 usec to 2157 usec for pid 979 (zabbix_server) calcru: runtime went backwards from 2179 usec to 1542 usec for pid 972 (zabbix_server) calcru: runtime went backwards from 1833 usec to 927 usec for pid 1026 (cron) calcru: runtime went backwards from 1091 usec to 551 usec for pid 1022 (sendmail) calcru: runtime went backwards from 823 usec to 416 usec for pid 1022 (sendmail) calcru: runtime went backwards from 4691 usec to 2735 usec for pid 1011 (sendmail) calcru: runtime went backwards from 1162 usec to 587 usec for pid 1011 (sendmail) calcru: runtime went backwards from 947 usec to 478 usec for pid 1010 (httpd) calcru: runtime went backwards from 895 usec to 452 usec for pid 1009 (httpd) calcru: runtime went backwards from 954 usec to 482 usec for pid 1008 (httpd) calcru: runtime went backwards from 909 usec to 459 usec for pid 1007 (httpd) calcru: runtime went backwards from 1214 usec to 653 usec for pid 1006 (httpd) calcru: runtime went backwards from 201793 usec to 102672 usec for pid 996 (httpd) calcru: runtime went backwards from 1105 usec to 1064 usec for pid 991 (sshd) calcru: runtime went backwards from 45196 usec to 23221 usec for pid 974 (zabbix_server) calcru: runtime went backwards from 2192 usec to 1436 usec for pid 973 (zabbix_server) calcru: runtime went backwards from 605 usec to 306 usec for pid 968 (zabbix_server) calcru: runtime went backwards from 626 usec to 316 usec for pid 967 (zabbix_server) calcru: runtime went backwards from 773 usec to 391 usec for pid 966 (zabbix_server) calcru: runtime went backwards from 663 usec to 335 usec for pid 965 (zabbix_server) calcru: runtime went backwards from 799 usec to 404 usec for pid 964 (zabbix_server) calcru: runtime went backwards from 37723 usec to 19422 usec for pid 963 (zabbix_server) calcru: runtime went backwards from 49425 usec to 26265 usec for pid 962 (zabbix_server) calcru: runtime went backwards from 38883 usec to 20993 usec for pid 961 (zabbix_server) calcru: runtime went backwards from 53478 usec to 27411 usec for pid 960 (zabbix_server) calcru: runtime went backwards from 40845 usec to 21750 usec for pid 959 (zabbix_server) calcru: runtime went backwards from 47835 usec to 24455 usec for pid 958 (zabbix_server) calcru: runtime went backwards from 2257 usec to 1612 usec for pid 957 (zabbix_server) calcru: runtime went backwards from 2361 usec to 1193 usec for pid 932 (zabbix_agentd) calcru: runtime went backwards from 32053 usec to 16204 usec for pid 928 (zabbix_server) calcru: runtime went backwards from 32517 usec to 19280 usec for pid 697 (sh) calcru: runtime went backwards from 1128021 usec to 868022 usec for pid 697 (sh) calcru: runtime went backwards from 9609 usec to 5601 usec for pid 601 (syslogd) calcru: runtime went backwards from 688 usec to 348 usec for pid 106 (adjkerntz) calcru: runtime went backwards from 12 usec to 6 usec for pid 3 (sctp_iterator) calcru: runtime went backwards from 5467 usec to 2853 usec for pid 2 (fdc0) calcru: runtime went backwards from 11306 usec to 8408 usec for pid 14 (rand_harvestq) calcru: runtime went backwards from 136242 usec to 76230 usec for pid 13 (geom) calcru: runtime went backwards from 156117 usec to 118882 usec for pid 12 (intr) calcru: runtime went backwards from 7820 usec to 4847 usec for pid 1 (init) calcru: runtime went backwards from 2340313 usec to 1226503 usec for pid 1 (init) calcru: runtime went backwards from 4114 usec to 2164 usec for pid 0 (kernel) Is it bad? Or I have to ignore them? I'm not talking authoratatively but in some virtualization systems, if the TSCs on the various processors are not sufficiently synchronized, then when the hypervisor switches you from one CPU to another, you can get odd (sometimes negative) time readings which give messages like
Hyper-V
Hi! I upgraded to 10.0-BETA3 but still see messages like Tue Nov 5 11:09:05 MSK 2013 calcru: runtime went backwards from 1728 usec to 889 usec for pid 981 (zabbix_server) calcru: runtime went backwards from 1229 usec to 703 usec for pid 976 (zabbix_server) calcru: runtime went backwards from 1019 usec to 595 usec for pid 978 (zabbix_server) calcru: runtime went backwards from 3041 usec to 1611 usec for pid 969 (zabbix_server) calcru: runtime went backwards from 1078 usec to 595 usec for pid 975 (zabbix_server) calcru: runtime went backwards from 1114 usec to 617 usec for pid 980 (zabbix_server) calcru: runtime went backwards from 1086 usec to 602 usec for pid 977 (zabbix_server) calcru: runtime went backwards from 1195 usec to 782 usec for pid 970 (zabbix_server) calcru: runtime went backwards from 3206 usec to 2157 usec for pid 979 (zabbix_server) calcru: runtime went backwards from 2179 usec to 1542 usec for pid 972 (zabbix_server) calcru: runtime went backwards from 1833 usec to 927 usec for pid 1026 (cron) calcru: runtime went backwards from 1091 usec to 551 usec for pid 1022 (sendmail) calcru: runtime went backwards from 823 usec to 416 usec for pid 1022 (sendmail) calcru: runtime went backwards from 4691 usec to 2735 usec for pid 1011 (sendmail) calcru: runtime went backwards from 1162 usec to 587 usec for pid 1011 (sendmail) calcru: runtime went backwards from 947 usec to 478 usec for pid 1010 (httpd) calcru: runtime went backwards from 895 usec to 452 usec for pid 1009 (httpd) calcru: runtime went backwards from 954 usec to 482 usec for pid 1008 (httpd) calcru: runtime went backwards from 909 usec to 459 usec for pid 1007 (httpd) calcru: runtime went backwards from 1214 usec to 653 usec for pid 1006 (httpd) calcru: runtime went backwards from 201793 usec to 102672 usec for pid 996 (httpd) calcru: runtime went backwards from 1105 usec to 1064 usec for pid 991 (sshd) calcru: runtime went backwards from 45196 usec to 23221 usec for pid 974 (zabbix_server) calcru: runtime went backwards from 2192 usec to 1436 usec for pid 973 (zabbix_server) calcru: runtime went backwards from 605 usec to 306 usec for pid 968 (zabbix_server) calcru: runtime went backwards from 626 usec to 316 usec for pid 967 (zabbix_server) calcru: runtime went backwards from 773 usec to 391 usec for pid 966 (zabbix_server) calcru: runtime went backwards from 663 usec to 335 usec for pid 965 (zabbix_server) calcru: runtime went backwards from 799 usec to 404 usec for pid 964 (zabbix_server) calcru: runtime went backwards from 37723 usec to 19422 usec for pid 963 (zabbix_server) calcru: runtime went backwards from 49425 usec to 26265 usec for pid 962 (zabbix_server) calcru: runtime went backwards from 38883 usec to 20993 usec for pid 961 (zabbix_server) calcru: runtime went backwards from 53478 usec to 27411 usec for pid 960 (zabbix_server) calcru: runtime went backwards from 40845 usec to 21750 usec for pid 959 (zabbix_server) calcru: runtime went backwards from 47835 usec to 24455 usec for pid 958 (zabbix_server) calcru: runtime went backwards from 2257 usec to 1612 usec for pid 957 (zabbix_server) calcru: runtime went backwards from 2361 usec to 1193 usec for pid 932 (zabbix_agentd) calcru: runtime went backwards from 32053 usec to 16204 usec for pid 928 (zabbix_server) calcru: runtime went backwards from 32517 usec to 19280 usec for pid 697 (sh) calcru: runtime went backwards from 1128021 usec to 868022 usec for pid 697 (sh) calcru: runtime went backwards from 9609 usec to 5601 usec for pid 601 (syslogd) calcru: runtime went backwards from 688 usec to 348 usec for pid 106 (adjkerntz) calcru: runtime went backwards from 12 usec to 6 usec for pid 3 (sctp_iterator) calcru: runtime went backwards from 5467 usec to 2853 usec for pid 2 (fdc0) calcru: runtime went backwards from 11306 usec to 8408 usec for pid 14 (rand_harvestq) calcru: runtime went backwards from 136242 usec to 76230 usec for pid 13 (geom) calcru: runtime went backwards from 156117 usec to 118882 usec for pid 12 (intr) calcru: runtime went backwards from 7820 usec to 4847 usec for pid 1 (init) calcru: runtime went backwards from 2340313 usec to 1226503 usec for pid 1 (init) calcru: runtime went backwards from 4114 usec to 2164 usec for pid 0 (kernel) Is it bad? Or I have to ignore them? ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
Re: Hyper-V driver development status
So, is it going to be included in FreeBSD 10 now? As I can see code freeze for CURRENT is planned on 7 September 2013 (10 days from now). http://www.freebsd.org/releases/10.0R/schedule.html It's not much time. 2013/8/17 Abhishek Gupta (LIS) abgu...@microsoft.com: Hi Pavel, Thanks for inquiring about the BSD drivers for Hyper-V. We (Microsoft) are actively working on these drivers and the intent is to get them integrated in to the kernel within the next month or so. There are some issues we are trying to address: a) Witness panic during boot time due to a specific memory allocation. b) Bypass usage of disk labels to ensure that devices do not get reordered under FreeBSD. c) Locks up on 32 bit builds. d) Lack of KVP support for BSD drivers We are working hard every day to finish these items. Even though an official announcement has not gone out, I would highly encourage you to start testing the drivers. As they are in project branch now they should be easy to compile. The only requirement is to set up disk labels before rebooting a kernel that includes the Hyper-V drivers otherwise the system will not boot up. Please let me know if you have more questions or have feedback. Thanks again for your interest. Abhishek From: Peter Grehan gre...@freebsd.org Sent: Friday, August 16, 2013 11:30 AM To: Abhishek Gupta (LIS) Subject: Fwd: Hyper-V driver development status Want to reply to this ? Original Message Subject: Hyper-V driver development status Date: Fri, 16 Aug 2013 17:00:57 +0400 From: Pavel Timofeev tim...@gmail.com To: freebsd-virtualization@freebsd.org Hi all! I'd like to know what's happening with hyper-v driver development in FreeBSD. I know about svn branch but there is no active progress in there. What's the roadmap? Or we are waiting for something? What is the current status of that project? I could do some tests. Or it's too early? I'm looking forward to use it =) ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to freebsd-virtualization-unsubscr...@freebsd.org
