From KVM
Line 1: Line 1:
 
= KVM-Autotest TODO list =
 
= KVM-Autotest TODO list =
 +
 +
== Network Patchset ==
 +
 +
Problems with nw patchset
 +
 +
1) multicast
 +
kvm.virtio_blk.smp2.Fedora.13.64.virtio_net.multicast Ping return non-zero value PING 225.0.0.1 (225.0.0.1) from 10.16.72.124 virtio_0_5900: 56(84) bytes of data.
 +
 +
2) netperf
 +
17:35:11 DEBUG| Execute netperf client test: /root/autotest/client/tests/netperf2/netperf-2.4.5/src/netperf -t TCP_CRR -H  10.16.74.142 -l 60 -- -m 1
 +
17:35:45 ERROR| Fail to execute netperf test, protocol:TCP_CRR
 +
17:35:45 DEBUG| Execute netperf client test: /root/autotest/client/tests/netperf2/netperf-2.4.5/src/netperf -t UDP_RR -H 10.16.74.142 -l 60 -- -m 1
 +
17:36:06 ERROR| Fail to execute netperf test, protocol:UDP_RR
 +
 +
3) vlan
 +
TestError: Fail to configure ip for eth0.1
 +
17:46:27 DEBUG| Sending command: ifconfig eth0.1 192.168.1.1
 +
17:46:27 DEBUG| Command failed; status: 255, output: SIOCSIFADDR: No such device eth0.1: unknown interface: No such device
 +
17:46:27 DEBUG| Sending command: if [[ -e /proc/net/vlan/eth0.1 ]];then vconfig rem eth0.1;fi
 +
17:46:27 DEBUG| Sending command: if [[ -e /proc/net/vlan/eth0.1 ]];then vconfig rem eth0.1;fi
 +
17:46:28 INFO | rem eth0.1
 +
17:46:28 DEBUG| Sending command: if [[ -e /proc/net/vlan/eth0.2 ]];then vconfig rem eth0.2;fi
 +
17:46:28 DEBUG| Sending command: if [[ -e /proc/net/vlan/eth0.2 ]];then vconfig rem eth0.2;fi
 +
17:46:28 INFO | rem eth0.2
 +
17:46:28 DEBUG| Sending command: if [[ -e /proc/net/vlan/eth0.3 ]];then vconfig rem eth0.3;fi
 +
17:46:28 DEBUG| Sending command: if [[ -e /proc/net/vlan/eth0.3 ]];then vconfig rem eth0.3;fi
 +
17:46:28 INFO | rem eth0.3
 +
17:46:28 DEBUG| Sending command: if [[ -e /proc/net/vlan/eth0.4 ]];then vconfig rem eth0.4;fi
 +
17:46:29 DEBUG| Sending command: if [[ -e /proc/net/vlan/eth0.4 ]];then vconfig rem eth0.4;fi
 +
17:46:29 INFO | rem eth0.4
 +
17:46:29 DEBUG| Sending command: if [[ -e /proc/net/vlan/eth0.5 ]];then vconfig rem eth0.5;fi
 +
17:46:29 DEBUG| Sending command: if [[ -e /proc/net/vlan/eth0.5 ]];then vconfig rem eth0.5;fi
 +
17:46:29 INFO | rem eth0.5
 +
17:46:29 ERROR| Test failed: TestError: Fail to configure ip for eth0.1
 +
 +
4) ethtool
 +
Address comment from mst:
 +
 +
> > Initialize the callbacks first and execute all the sub
 +
> > tests one by one, all the result will be check at the
 +
> > end. When execute this test, vhost should be enabled,
 +
> > then most of new features can be used. Vhost doesn't
 +
> > support VIRTIO_NET_F_MRG_RXBUF, so do not check large
 +
> > packets in received offload test.
 +
 +
Well, it does support that now, in any case VIRTIO_NET_F_MRG_RXBUF
 +
is not required for large packets: it's an optimization saving
 +
guest memory, really. So no need to special-case.
  
 
== Framework ==
 
== Framework ==

Revision as of 21:56, 29 September 2010

KVM-Autotest TODO list

Network Patchset

Problems with nw patchset

1) multicast

kvm.virtio_blk.smp2.Fedora.13.64.virtio_net.multicast Ping return non-zero value PING 225.0.0.1 (225.0.0.1) from 10.16.72.124 virtio_0_5900: 56(84) bytes of data.

2) netperf

17:35:11 DEBUG| Execute netperf client test: /root/autotest/client/tests/netperf2/netperf-2.4.5/src/netperf -t TCP_CRR -H   10.16.74.142 -l 60 -- -m 1
17:35:45 ERROR| Fail to execute netperf test, protocol:TCP_CRR
17:35:45 DEBUG| Execute netperf client test: /root/autotest/client/tests/netperf2/netperf-2.4.5/src/netperf -t UDP_RR -H 10.16.74.142 -l 60 -- -m 1
17:36:06 ERROR| Fail to execute netperf test, protocol:UDP_RR

3) vlan

TestError: Fail to configure ip for eth0.1
17:46:27 DEBUG| Sending command: ifconfig eth0.1 192.168.1.1
17:46:27 DEBUG| Command failed; status: 255, output: SIOCSIFADDR: No such device eth0.1: unknown interface: No such device
17:46:27 DEBUG| Sending command: if  -e /proc/net/vlan/eth0.1 ;then vconfig rem eth0.1;fi
17:46:27 DEBUG| Sending command: if  -e /proc/net/vlan/eth0.1 ;then vconfig rem eth0.1;fi
17:46:28 INFO | rem eth0.1
17:46:28 DEBUG| Sending command: if  -e /proc/net/vlan/eth0.2 ;then vconfig rem eth0.2;fi
17:46:28 DEBUG| Sending command: if  -e /proc/net/vlan/eth0.2 ;then vconfig rem eth0.2;fi
17:46:28 INFO | rem eth0.2
17:46:28 DEBUG| Sending command: if  -e /proc/net/vlan/eth0.3 ;then vconfig rem eth0.3;fi
17:46:28 DEBUG| Sending command: if  -e /proc/net/vlan/eth0.3 ;then vconfig rem eth0.3;fi
17:46:28 INFO | rem eth0.3
17:46:28 DEBUG| Sending command: if  -e /proc/net/vlan/eth0.4 ;then vconfig rem eth0.4;fi
17:46:29 DEBUG| Sending command: if  -e /proc/net/vlan/eth0.4 ;then vconfig rem eth0.4;fi
17:46:29 INFO | rem eth0.4
17:46:29 DEBUG| Sending command: if  -e /proc/net/vlan/eth0.5 ;then vconfig rem eth0.5;fi
17:46:29 DEBUG| Sending command: if  -e /proc/net/vlan/eth0.5 ;then vconfig rem eth0.5;fi
17:46:29 INFO | rem eth0.5
17:46:29 ERROR| Test failed: TestError: Fail to configure ip for eth0.1

4) ethtool Address comment from mst:

> > Initialize the callbacks first and execute all the sub
> > tests one by one, all the result will be check at the
> > end. When execute this test, vhost should be enabled,
> > then most of new features can be used. Vhost doesn't
> > support VIRTIO_NET_F_MRG_RXBUF, so do not check large
> > packets in received offload test.
Well, it does support that now, in any case VIRTIO_NET_F_MRG_RXBUF
is not required for large packets: it's an optimization saving
guest memory, really. So no need to special-case.

Framework

  • Fix problem with serial_login() - probably we forgot to setup the guest in a way that we hook up a getty in the serial port.
  • Add a BLOCKED status on the autotest database, that tells the user that some feature testing was left out due to a problem on its dependencies.
  • Eventually get rid of all dependency on slirp/userspace networking to do any feature testing, since it is not a supported feature. Slipstreaming the kickstart in the install CD is a possibility, and it ends up being simpler than configuring an internal network thing for testing.
  • Make it possible to install virtio nw driver under windows xp/2003 guests without having to resort to the msi installer - assignee? Considerations: We would like to avoid slipstreaming the drivers in the iso cdrom, since we feel it is too messy. If after honest try we can't find another way, let's give slipstreaming a try.
  • Re-schedule the unittest job on the autotest server. DONE
  • Multi host migration - 1 week to recap the patch that was sent by QA, improve on it, have an updated patch by then - lmr

Deadline: 09/15 - Could not meet it, has to postpone New tentative deadline: 09/22

  • Libvirt integration - 1 week coordinating what it needs to be done, another 4 weeks (at least) implementing it - QA, lmr, mgoldish, help from libvirt developers

Deadline: 09/15 - Could not meet it, has to postpone New tentative deadline: 09/22

  • Include support for host kernel install on the KVM default control file - Using autotest standard API to get it done - lmr

Tentative deadline: 09/22

  • Enable "guest-load" for a VMs before/while tests are running (e.g. migration of a VM, while a movie is playing on it) [Michael]
  • Add a way to shutdown the VMs when the whole job completes (compared to doing nothing or shutting-down after every test) [Michael]
  • Random generated macaddress and ifname (BeiJing QE)


Client Side Test

  • Using dnsmasq in unattended_install to replace userspace network with private bridge [Amos] status: have implemented in internal tree
  • Bonding test [Amos]
  • If needed, write a netperf/iperf test [jasowang] [NEEDS MORE WORK]
  • pxe boot * net types [BeiJing QE]
  • Further migration [BeiJing QE]
  • Multiple nics [BeiJing QE]
  • multiple disks [BeiJing QE]
  • -vga stg, nographics
  • Verify SMBIOS/DMI data (UUID, for example) [Yolkfull]
  • Disk serial number (for IDE, SCSI, VirtIO drives) [Yolkfull]
  • Test block device cancellation path using device mapper to generate errors ( after we had a crash in de_write_dma_cb)
  • Extend pci_assignable to support other PCI devices (USB, video cards, TV card etc)
  • Different CPU flags support [BeiJing QE]
  • Use private bridge and dnsmasq to do the unattended installation [BeiJing QE]
  • Passthrough the perf keys of run_autotest to autotest server [BeiJing QE]

Server Side Tests

  • Run netperf test between two guests [Yolkfull]
  • Migration with/without workload(dbench,lmbench etc) [Yolkfull]
  • Register the virtual machine into autotest server and run benchmark through autotest server [BeiJing QE]

Misc

  • Add docstrings to all functions that still lack them, including the ones in stepmaker.py, stepeditor.py and kvm_tests.py [Michael]
  • Document the setupssh.iso and setuptelnet.iso creation procedures in the wiki
  • Rename all Windows ISOs currently used to their official MSDN names
  • Fill the sections "Working with step files" and "Step file creation tips" in the wiki [Michael]

Bugs