Re: [ovirt-users] multiple ip routing table issue
On Tue, Nov 21, 2017 at 1:24 AM, Edward Clay wrote: > Hello, > > We have an issue where hosts are configured with the public facing nework > interface as the ovirtmgmt network and it's default route is added to a > ovirt created table but not to the main routing table. From my searching > I've found this snippet from https://www.ovirt.org/develop/ > release-management/features/network/multiple-gateways/ which seems to > explain why I can't ping anything or communicate with any other system > needing a default route. > By default, the default route is set on the ovirtmgmt network (the default one, defined on the interface/ip which you added the host to Engine). Do you have a different network set up which you will like to set the default route on? > > "And finally, here's the host's main routing table. Any traffic coming in > to the host will use the ip rules and an interface's routing table. The > main routing table is only used for traffic originating from the host." > > I'm seeing the following main and custom ovirt created tables. > > main: > # ip route show table main > 10.0.0.0/8 via 10.4.16.1 dev enp3s0.106 > 10.4.16.0/24 dev enp3s0.106 proto kernel scope link src 10.4.16.15 > 1.1.1.0/24 dev PUBLICB proto kernel scope link src 1.1.1.1 169.254.0.0/16 > dev enp6s0 scope link metric 1002 > 169.254.0.0/16 dev enp3s0 scope link metric 1003 > 169.254.0.0/16 dev enp7s0 scope link metric 1004 > 169.254.0.0/16 dev enp3s0.106 scope link metric 1020 > 169.254.0.0/16 dev PRIVATE scope link metric 1022 > 169.254.0.0/16 dev PUBLIC scope link metric 1024 > > table 1138027711 > # ip route show table 1138027711 > default via 1.1.1.1 dev PUBLIC > 1.1.1.0/24 via 1.1.1.1 dev PUBLIC > > If I manually execute the following command to add the default route as > well to the main table I can ping ouside of the local network. > > ip route add 0.0.0.0/0 via 1.1.1.1 dev PUBLIC > > If I attempt to modify the /etc/sysconfig/network-scripts/route-PUBLIC ad > reboot the server ad one would think this file is recreated by vdsm on boot. > > What I'm looking for is the correct way to setup a default gateway for the > main routing table so the hosts can get OS updates and communicate with the > outside world. > Providing the output from "ip addr" may help clear up some things. It looks like you have on the host the default route set as 10.4.16.1 (on enp3s0.106), could you elaborate what this interface is? Thanks, Edy. > ___ > Users mailing list > Users@ovirt.org > http://lists.ovirt.org/mailman/listinfo/users > > ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
[ovirt-users] multiple ip routing table issue
Hello, We have an issue where hosts are configured with the public facing nework interface as the ovirtmgmt network and it's default route is added to a ovirt created table but not to the main routing table. From my searching I've found this snippet from https://www.ovirt.org/develop /release-management/features/network/multiple-gateways/ which seems to explain why I can't ping anything or communicate with any other system needing a default route. "And finally, here's the host's main routing table. Any traffic coming in to the host will use the ip rules and an interface's routing table. The main routing table is only used for traffic originating from the host." I'm seeing the following main and custom ovirt created tables. main: # ip route show table main 10.0.0.0/8 via 10.4.16.1 dev enp3s0.106 10.4.16.0/24 dev enp3s0.106 proto kernel scope link src 10.4.16.15 1.1.1.0/24 dev PUBLICB proto kernel scope link src 1.1.1.1 169.254.0.0/16 dev enp6s0 scope link metric 1002 169.254.0.0/16 dev enp3s0 scope link metric 1003 169.254.0.0/16 dev enp7s0 scope link metric 1004 169.254.0.0/16 dev enp3s0.106 scope link metric 1020 169.254.0.0/16 dev PRIVATE scope link metric 1022 169.254.0.0/16 dev PUBLIC scope link metric 1024 table 1138027711 # ip route show table 1138027711 default via 1.1.1.1 dev PUBLIC 1.1.1.0/24 via 1.1.1.1 dev PUBLIC If I manually execute the following command to add the default route as well to the main table I can ping ouside of the local network. ip route add 0.0.0.0/0 via 1.1.1.1 dev PUBLIC If I attempt to modify the /etc/sysconfig/network-scripts/route-PUBLIC ad reboot the server ad one would think this file is recreated by vdsm on boot. What I'm looking for is the correct way to setup a default gateway for the main routing table so the hosts can get OS updates and communicate with the outside world.___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [ovirt-users] Why was HE Install made OVA only?
On Mon, Nov 20, 2017 at 4:39 PM, Alan Griffiths wrote: > Hi, > > What was the reasoning behind making Hosted Engine install OVA only? > The PXEBoot feature always worked really well for me, and now I have a > number of extra steps to achieve the same end result. > > Do you mean that you were customizing the image shipped via PXE? Deploying from the OVA is pretty convenient, if you want just to forget about it, you have to install also ovirt-engine-appliance rpm when you install ovirt-hosted-engine-setup one. I don't see other additional steps. > Thanks, > > Alan > ___ > Users mailing list > Users@ovirt.org > http://lists.ovirt.org/mailman/listinfo/users > ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [ovirt-users] Openshift in VMs created from oVirt
Thanks Evgheni for your message. Minishift/Minikube require to prepare the VM (boot2docker, docker machine, drivers, etc.) before running Openshift/Kubernetes inside. Anyway, I'm going to explore these suggestions. 1) oVirt templates info: https://www.ovirt.org/documentation/vmm-guide/chap-Templates 2) Is this (https://cloud-init.io) Cloud Init? 3) Does that mean executing the command (minishift start ..) in the VM from oVirt right after the VM has been created?. Do you have any info about this?. I appreciate a lot your support. Regards. -- wr On 20 November 2017 at 11:48, Evgheni Dereveanchin wrote: > Hi Wayna, > > A VM with OpenShift is no different from any other VM. Users should be > able to reach it via SSH if your network/firewall settings allow that. If > you want to auto-provision minishift I see several options: > 1) install minishift manually in a VM and save it as a template to > provision new VMs from > 2) use a cloud image from glance and invoke cloud-init to run the installer > 3) deploy new VMs using your standard workflow, then use Ansible or other > configuration management tools to setup minishift after install. > > Hope this helps. > > On Fri, Nov 17, 2017 at 2:26 PM, Wayna Runa wrote: > >> Hi there!, >> >> I've oVirt running and providing several VMs (CentOS and RedHat), now I >> want to provide VMs with minimalist Openshift cluster running inside to >> developers. >> Minishift creates a VM with Openshift locally, that is fine, but now we >> have oVirt to provide VMs which the developers can get access remotely. >> >> How I can use oVirt to do that?. >> Thanks in advance. >> >> -- >> *Wayna Runa* >> >> ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
[ovirt-users] Why was HE Install made OVA only?
Hi, What was the reasoning behind making Hosted Engine install OVA only? The PXEBoot feature always worked really well for me, and now I have a number of extra steps to achieve the same end result. Thanks, Alan ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
[ovirt-users] slow performance with export storage on glusterfs
Hi, I am trying realize why is exporting of vm to export storage on glusterfs such slow. I am using oVirt and RHV, both instalations on version 4.1.7. Hosts have dedicated nics for rhevm network - 1gbps, data storage itself is on FC. GlusterFS cluster lives separate on 4 dedicated hosts. It has slow disks but I can achieve about 200-400mbit throughput in other applications (we are using it for "cold" data, backups mostly). I am using this glusterfs cluster as backend for export storage. When I am exporting vm I can see only about 60-80mbit throughput. What could be the bottleneck here? Could it be qemu-img utility? vdsm 97739 0.3 0.0 354212 29148 ?S smime.p7s Description: S/MIME Cryptographic Signature ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [ovirt-users] oVirt 4.1.7: Duplicate MAC addresses had to be introduced into mac pool violating no duplicates setting.
Hi Alona, El 2017-11-19 08:31, Alona Kaplan escribió: Hi Nicolas, There was a bug in oVirt [1] that in some cases a mac pool that doesn't support duplicate macs could get duplicates. The bug was (hopefully) fixed. The error that you see means that you have a mac pool that was affected by this bug and have duplicate macs although it doesn't support it. The second part of the error should contain the list of the duplicate macs ("Following MACs violates duplicity restriction, and was pushed into MAC pool without respect to it:"). Indeed, being this message: engine.log-20171117.gz:2017-11-16 10:34:55,373Z ERROR [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (ServerService Thread Pool -- 52) [] EVENT_ID: MAC_ADDRESS_VIOLATES_NO_DUPLICATES_SETTING(10,917), Following MACs violates duplicity restriction, and was pushed into MAC pool without respect to it:[00:1a:4a:97:5f:16, 00:1a:4a:97:5f:05, 00:1a:4a:97:5f:1d, 00:1a:4a:97:5f:07, 00:1a:4a:97:5f:0f, 00:1a:4a:4d:cc:1c, 00:1a:4a:97:5f:15, 00:1a:4a:97:5f:0d] It is just a warning, you can ignore it. But we advise you to go over the duplicate macs and fix them. However, recently I made a small script to find duplicate MACs because I was having an issue with dupe MACs (which BTW you helped me to debug in BZ [1]). I'm attaching the script just in case something is wrong in it, but the script shows no dupe MAC addresses currently. Maybe in the past (before the upgrade) there was dupes with those MAC addresses, but not now. The output of the script shows only one machine per MAC address: 00:1a:4a:97:5f:16 ['VM1'] 00:1a:4a:97:5f:05 ['VM2'] 00:1a:4a:97:5f:1d ['VM3'] 00:1a:4a:97:5f:07 ['VM4'] 00:1a:4a:97:5f:0f ['VM5'] 00:1a:4a:4d:cc:1c ['VM6'] 00:1a:4a:97:5f:15 ['VM7'] 00:1a:4a:97:5f:0d ['VM8'] Should I file a bug or simply ignore it? Thanks. [1]: https://bugzilla.redhat.com/show_bug.cgi?id=1497242 Alona. [1] https://bugzilla.redhat.com/1485688 [2] On Fri, Nov 17, 2017 at 4:54 PM, wrote: Hi, We just upgraded from 4.1.6 to 4.1.7 and just after the engine was brought back the following event showed up: Duplicate MAC addresses had to be introduced into mac pool violating no duplicates setting. Can someone explain why is that? Thanks. Nicolás ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users [1] Links: -- [1] http://lists.ovirt.org/mailman/listinfo/users [2] https://bugzilla.redhat.com/1485688#!/usr/bin/env python from sys import path import os.path from sys import exit from ovirtsdk4 import Connection, types from scriptsparams import URI, USERNAME, PASSWORD, CERTPATH conn = Connection( url=URI, username=USERNAME, password=PASSWORD, ca_file=CERTPATH ) if not conn.test(raise_exception=False): print "ERROR: Cannot connect. Is engine up? Are credentials ok?" exit(1) vms_macs = {} sys_serv = conn.system_service() vms_serv = sys_serv.vms_service() macs = [] for vm in vms_serv.list(): nics = conn.follow_link(vm.nics) for nic in nics: if nic.mac.address in vms_macs: vms_macs[nic.mac.address].append(vm.name) else: vms_macs[nic.mac.address] = [vm.name] macs.append(nic.mac.address) print '00:1a:4a:97:5f:16', vms_macs['00:1a:4a:97:5f:16'] print '00:1a:4a:97:5f:05', vms_macs['00:1a:4a:97:5f:05'] print '00:1a:4a:97:5f:1d', vms_macs['00:1a:4a:97:5f:1d'] print '00:1a:4a:97:5f:07', vms_macs['00:1a:4a:97:5f:07'] print '00:1a:4a:97:5f:0f', vms_macs['00:1a:4a:97:5f:0f'] print '00:1a:4a:4d:cc:1c', vms_macs['00:1a:4a:4d:cc:1c'] print '00:1a:4a:97:5f:15', vms_macs['00:1a:4a:97:5f:15'] print '00:1a:4a:97:5f:0d', vms_macs['00:1a:4a:97:5f:0d'] ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [ovirt-users] Some tests results: lustrefs over nfs on VM
On Mon, Nov 20, 2017 at 12:23 PM, Yaniv Kaul wrote: > > > Define QoS on the NIC. > But I think you wish to limit IO, no? > Y. > For the moment QoS is unlimited. Actually for some tasks I wish to allocate 80% of 10Gbit interface, but the VM interface is always 1Gbit. Inside the QoS of the host interface I cannot put 8000 for the Rate Limit, it claims that rate limit should be between number 1...1024, looks like it assumes only 1Gbit interfaces? a. ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [ovirt-users] Openshift in VMs created from oVirt
Hi Wayna, A VM with OpenShift is no different from any other VM. Users should be able to reach it via SSH if your network/firewall settings allow that. If you want to auto-provision minishift I see several options: 1) install minishift manually in a VM and save it as a template to provision new VMs from 2) use a cloud image from glance and invoke cloud-init to run the installer 3) deploy new VMs using your standard workflow, then use Ansible or other configuration management tools to setup minishift after install. Hope this helps. On Fri, Nov 17, 2017 at 2:26 PM, Wayna Runa wrote: > Hi there!, > > I've oVirt running and providing several VMs (CentOS and RedHat), now I > want to provide VMs with minimalist Openshift cluster running inside to > developers. > Minishift creates a VM with Openshift locally, that is fine, but now we > have oVirt to provide VMs which the developers can get access remotely. > > How I can use oVirt to do that?. > Thanks in advance. > > -- > *Wayna Runa* > > > ___ > Users mailing list > Users@ovirt.org > http://lists.ovirt.org/mailman/listinfo/users > > -- Regards, Evgheni Dereveanchin ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [ovirt-users] Some tests results: lustrefs over nfs on VM
On Nov 19, 2017 11:05 PM, "Arman Khalatyan" wrote: hi Yaniv, yes for sure I hit some cache in between but not the vm cache, it has a 4GB ram, with oflag=direct I get about 120MB/s for the data analysis the cache is our friend:) The backend is a lustrefs 2.10.x. yes we have dedicated 10G on the hosts, where we can limit the vm interface to 10Gbit? Define QoS on the NIC. But I think you wish to limit IO, no? Y. Am 19.11.2017 8:33 nachm. schrieb "Yaniv Kaul" : On Sun, Nov 19, 2017 at 7:08 PM, Arman Khalatyan wrote: > Hi, in our environment we got pretty good io performance on VM, with > following configuration: > lustrebox: /lust mounted on "GATEWAY" over IB > GATEWAY: export /lust as nfs4 on 10G interface > VM(test.vm): import as NFS over 10G interface > > [r...@test.vm ~]# dd if=/dev/zero bs=128K count=10 > Without oflag=direct, you are hitting (somewhat) the cache. > > of=/test/tmp/test.tmp > 10+0 records in > 10+0 records out > 1310720 <(310)%20720-> bytes (13 GB) copied, 20.8402 s, 629 MB/s > looks promising for the future deployments. > Very - what's the backend storage? > > only one problem remains that on heavy io I get some wornings that the vm > network is saturated, are there way to configure the bandwidth limits to > 10G for the VM Interface?? > Yes, but you really need a dedicated storage interface, no? Y. > > > thank you beforehand, > Arman. > > ___ > Users mailing list > Users@ovirt.org > http://lists.ovirt.org/mailman/listinfo/users > > ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [ovirt-users] VDSM command GetVmsInfoVDS failed: Missing OVF file from VM
Yes, you can remove it On Mon, Nov 20, 2017 at 8:10 AM, Алексей Максимов < aleksey.i.maksi...@yandex.ru> wrote: > I found an empty directory in the Export domain storage: > > # ls -la /rhev/data-center/mnt/fs01.my.dom-holding.com:_mnt_quadstor- > vv1_ovirt-vm-backup/3a514c90-e574-4282-b1ee-779602e35f24/ > master/vms/f4429fa5-76a2-45a7-ae3e-4d8955d4f1a6 > > total 16 > drwxr-xr-x. 2 vdsm kvm 4096 Nov 9 02:32 . > drwxr-xr-x. 106 vdsm kvm 12288 Nov 9 02:32 .. > > I can just remove this directory? > > 19.11.2017, 18:51, "Benny Zlotnik" : > > + ovirt-users > > On Sun, Nov 19, 2017 at 5:40 PM, Benny Zlotnik > wrote: > > Hi, > > There are a couple of issues here, can you please open a bug so we can > track this properly? https://bugzilla.redhat.com/ > and attach all relevant logs > > I went over the logs, are you sure the export domain was formatted > properly? Couldn't find it in the engine.log > Looking at the logs it seems VMs were found on the export domain > (id=3a514c90-e574-4282-b1ee-779602e35f24) > > 2017-11-19 13:18:13,007+0300 INFO (jsonrpc/2) [storage.StorageDomain] > vmList=[u'01a4f53e-699e-4ea5-aef4-458638f23ce9', > u'03c9e965-710d-4fc8-be06-583abbd1d7a9', > u'07dab4f6-d677-4faa-9875-97bd6d601f49', > u'0b94a559-b31a-475d-9599-36e0dbea579a', > u'13b42f3a-3057-4eb1-ad4b-f4e52f6ff196', > u'151a4e75-d67a-4603-8f52-abfb46cb74c1', > u'177479f5-2ed8-4b6c-9120-ec067d1a1247', > u'18945b31-3ba5-4e54-9bf0-8fdc3a7d7411', > u'1e72be16-f540-4cfd-b0e9-52b66220a98b', > u'1ec85134-a7b5-46c2-9c6c-eaba340c5ffd', > u'20b88cfc-bfae-4983-8d83-ba4e0c7feeb7', > u'25fa96d1-6083-4daa-9755-026e632553d9', > u'273ffd05-6f93-4e4a-aac9-149360b5f0b4', > u'28188426-ae8b-4999-8e31-4c04fbba4dac', > u'28e9d5f2-4312-4d0b-9af9-ec1287bae643', > u'2b7093dc-5d16-4204-b211-5b3a1d729872', u'32ecfcbb-2678-4f43-8d59-418e > 03920693', u'3376ef0b-2af5-4a8b-9987-18f28f6bb334', > u'34d1150f-7899-44d9-b8cf-1c917822f624', > u'383bbfc6-6841-4476-b108-a1878ed9ce43', > u'388e372f-b0e8-408f-b21b-0a5c4a84c457', > u'39396196-42eb-4a27-9a57-a3e0dad8a361', > u'3fc02ca2-7a03-4d5e-bc21-688f138a914f', > u'4101ac1e-0582-4ebe-b4fb-c4aed39fadcf', > u'44e10588-8047-4734-81b3-6a98c229b637', > u'4794ca9c-5abd-4111-b19c-bdfbf7c39c86', > u'47a83986-d3b8-4905-b017-090276e967f5', > u'49d83471-a312-412e-b791-8ee0badccbb5', > u'4b1b9360-a48a-425b-9a2e-19197b167c99', > u'4d783e2a-2d81-435a-98c4-f7ed862e166b', > u'51976b6e-d93f-477e-a22b-0fa84400ff84', > u'56b77077-707c-4949-9ea9-3aca3ea912ec', > u'56dc5c41-6caf-435f-8146-6503ea3eaab9', > u'5729e036-5f6e-473b-9d1d-f1c4c5c55b2d', > u'5873f804-b992-4559-aff5-797f97bfebf7', > u'58b7a4ea-d572-4ab4-a4f1-55dddc5dc8e8', > u'590d1adb-52e4-4d29-af44-c9aa5d328186', > u'5c79f970-6e7b-4996-a2ce-1781c28bff79', > u'5feab1f2-9a3d-4870-a0f3-fd97ea3c85c3', > u'63749307-4486-4702-ade9-4324f5bfe80c', > u'6555ac11-7b20-4074-9d71-f86bc10c01f9', > u'66b4b8a0-b53b-40ea-87ab-75f6d9eef728', > u'672c4e12-628f-4dcd-a57e-b4ff822a19f3', > u'679c0445-512c-4988-8903-64c0c08b5fab', > u'6ae337d0-e6a0-489f-82e6-57a85f63176a', u'6d713cb9-993d-4822-a030-ac75 > 91794050', u'72a50ef0-945d-428a-a336-6447c4a70b99', > u'751dfefc-9e18-4f26-bed6-db412cdb258c', > u'7587db59-e840-41bc-96f3-b212b7b837a4', > u'778c969e-1d22-46e3-bdbe-e20e0c5bb967', > u'7810dec1-ee1c-4291-93f4-18e9a15fa8e2', > u'7a6cfe35-e493-4c04-8fc6-e0bc72efc72d', > u'7a7d814e-4586-40d5-9750-8896b00a6490', > u'7af76921-4cf2-4c3c-9055-59c24d9e8b08', > u'7d781e21-6613-41f4-bcea-8b57417e1211', > u'7da51499-d7db-49fd-88f6-bcac30e5dd86', > u'850a8041-77a4-4ae3-98f9-8d5f3a5778e6', > u'85169fe8-8198-492f-b988-b8e24822fd01', > u'87839926-8b84-482b-adec-5d99573edd9e', > u'8a7eb414-71fa-4f91-a906-d70f95ccf995', > u'8a9a1071-b005-4448-ba3f-c72bd7e0e34b', > u'8b73e593-8513-4a8e-b051-ce91765b22bd', > u'8cbd5615-4206-4e4a-992d-8705b2f2aac2', > u'92e9d966-c552-4cf9-b84a-21dda96f3f81', > u'95209226-a9a5-4ada-8eed-a672d58ba72c', > u'986ce2a5-9912-4069-bfa9-e28f7a17385d', > u'9f6c8d1d-da81-4020-92e5-1c14cf082d2c', > u'9ff87197-d089-4b2d-8822-b0d6f6e67292', > u'a0a0c756-fbe9-4f8e-b6e9-1f2d58f1d957', > u'a46d5615-8d9f-4944-9334-2fca2b53c27e', > u'a6a50244-366b-4b7c-b80f-04d7ce2d8912', > u'aa6a4de6-cc9e-4d79-a795-98326bbd83db', > u'accc0bc3-c501-4f0b-aeeb-6858f7e894fd', > u'b09e5783-6765-4514-a5a3-86e5e73b729b', > u'b1ecfe29-7563-44a9-b814-0faefac5465b', > u'baa542e1-492a-4b1b-9f54-e9566a4fe315', > u'bb91f9f5-98df-45b1-b8ca-9f67a92eef03', > u'bd11f11e-be3d-4456-917c-f93ba9a19abe', > u'bee3587e-50f4-44bc-a199-35b38a19ffc5', > u'bf573d58-1f49-48a9-968d-039e0916c973', > u'c01d466a-8ad8-4afe-b383-e365deebc6b8', > u'c0be5c12-be26-47b7-ad26-3ec2469f1d3f', > u'c31f4f53-c22b-40ff-8408-f36f591f55b5', > u'c530e339-99bf-48a2-a63a-cfd2a4dba198', > u'c8a610c8-72e5-4217-b4d9-130f85db1db7', > u'ca0567e1-d445-4875-94b1-85e31f331b87', > u'd2c3dab7-eb8c-410e-87c1-3d0139d9903c', > u'd330c4b6-bfa0-4061-aa1a-0f016598f2d0', > u'd5626b82-5079-4c42-84f5-fd972b493841', > u'd740e080-701d-4e07-becb-2b