From Dan Kenigsberg <danken(a)redhat.com>:
Dan Kenigsberg has submitted this change and it was merged.
Change subject: net: Small cleanup of _add_network and _del_network
......................................................................
net: Small cleanup of _add_network and _del_network
Removed unused arguments and made netinfo argument non optional.
This patch includes also some cleanup for config_network_test which had
to change to adjust to the changes.
Change-Id: I0a493b2c6312a04cd258415cba1a537a4df4bf95
Signed-off-by: Edward Haas <edwardh(a)redhat.com>
---
M lib/vdsm/network/legacy_switch.py
M lib/vdsm/network/netconfpersistence.py
M tests/network/config_network_test.py
3 files changed, 61 insertions(+), 71 deletions(-)
Approvals:
Jenkins CI: Passed CI tests
Petr Horáček: Looks good to me, but someone else must approve
Dan Kenigsberg: Looks good to me, approved
Edward Haas: Verified
--
To view, visit https://gerrit.ovirt.org/72652
To unsubscribe, visit https://gerrit.ovirt.org/settings
Gerrit-MessageType: merged
Gerrit-Change-Id: I0a493b2c6312a04cd258415cba1a537a4df4bf95
Gerrit-PatchSet: 4
Gerrit-Project: vdsm
Gerrit-Branch: master
Gerrit-Owner: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Dan Kenigsberg <danken(a)redhat.com>
Gerrit-Reviewer: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Jenkins CI
Gerrit-Reviewer: Petr Horáček <phoracek(a)redhat.com>
Gerrit-Reviewer: gerrit-hooks <automation(a)ovirt.org>
From Dan Kenigsberg <danken(a)redhat.com>:
Dan Kenigsberg has submitted this change and it was merged.
Change subject: net: Persist running config accross reboots
......................................................................
net: Persist running config accross reboots
Relocating running config from /var/run to /var/lib, making it persitent
accross reboots.
Running config represents a staging state, between the safe state
and the actual/current state (read mainly from the kernel).
As we have done so far, we assume that running config is in sync with
the actual state.
In future patches, we will aim to reduce running config content, leaving
only the information needed to identify which components are owned by
VDSM networking and possibly some additional metadata that cannot be
persisted on the device level.
Change-Id: I5fb2f53bbebb8faf57fededb15a2af8fa9dfc125
Signed-off-by: Edward Haas <edwardh(a)redhat.com>
---
M contrib/shell_helper
M lib/vdsm/network/netconfpersistence.py
M tests/functional/networkTests.py
M vdsm.spec.in
M vdsm/Makefile.am
M vdsm/vdsm-restore-net-config
M vdsm/vdsm-store-net-config.in
7 files changed, 16 insertions(+), 58 deletions(-)
Approvals:
Jenkins CI: Passed CI tests
Petr Horáček: Looks good to me, but someone else must approve
Dan Kenigsberg: Looks good to me, approved
Edward Haas: Verified
--
To view, visit https://gerrit.ovirt.org/72343
To unsubscribe, visit https://gerrit.ovirt.org/settings
Gerrit-MessageType: merged
Gerrit-Change-Id: I5fb2f53bbebb8faf57fededb15a2af8fa9dfc125
Gerrit-PatchSet: 6
Gerrit-Project: vdsm
Gerrit-Branch: master
Gerrit-Owner: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Dan Kenigsberg <danken(a)redhat.com>
Gerrit-Reviewer: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Jenkins CI
Gerrit-Reviewer: Petr Horáček <phoracek(a)redhat.com>
Gerrit-Reviewer: gerrit-hooks <automation(a)ovirt.org>
From Dan Kenigsberg <danken(a)redhat.com>:
Dan Kenigsberg has posted comments on this change.
Change subject: net: Small cleanup of _add_network and _del_network
......................................................................
Patch Set 4: Code-Review+2
raising
--
To view, visit https://gerrit.ovirt.org/72652
To unsubscribe, visit https://gerrit.ovirt.org/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: I0a493b2c6312a04cd258415cba1a537a4df4bf95
Gerrit-PatchSet: 4
Gerrit-Project: vdsm
Gerrit-Branch: master
Gerrit-Owner: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Dan Kenigsberg <danken(a)redhat.com>
Gerrit-Reviewer: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Jenkins CI
Gerrit-Reviewer: Petr Horáček <phoracek(a)redhat.com>
Gerrit-Reviewer: gerrit-hooks <automation(a)ovirt.org>
Gerrit-HasComments: No
From Dan Kenigsberg <danken(a)redhat.com>:
Dan Kenigsberg has posted comments on this change.
Change subject: net: Persist running config accross reboots
......................................................................
Patch Set 6: Code-Review+2
raising
--
To view, visit https://gerrit.ovirt.org/72343
To unsubscribe, visit https://gerrit.ovirt.org/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: I5fb2f53bbebb8faf57fededb15a2af8fa9dfc125
Gerrit-PatchSet: 6
Gerrit-Project: vdsm
Gerrit-Branch: master
Gerrit-Owner: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Dan Kenigsberg <danken(a)redhat.com>
Gerrit-Reviewer: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Jenkins CI
Gerrit-Reviewer: Petr Horáček <phoracek(a)redhat.com>
Gerrit-Reviewer: gerrit-hooks <automation(a)ovirt.org>
Gerrit-HasComments: No
From Dan Kenigsberg <danken(a)redhat.com>:
Dan Kenigsberg has submitted this change and it was merged.
Change subject: net: Modify restore-net to remove nets using setupNetworks
......................................................................
net: Modify restore-net to remove nets using setupNetworks
Overview:
The network restoration flow has a step in which acquired devices which
have been 'taken' by VDSM are restored.
When a device is acquired by VDSM, it is backed up, allowing later
rollback. The backed up data is cleared on setSafe or in case a
restoration is performed.
Problem:
When restoring in this step, an existing network (or partial network) is
removed by the underlying ifcfg configurator, but not through the proper
API.
Therefore, in such a scenario:
Clean host -> setupNetworks(add A) -> restore-net
(note that setSate() is missing)
Network A will be removed using ifcfg backup restoration flow without
issuing a setupNetworks(remove A).
One outcome of the missing proper removal of the network is that its
leftovers remain in the libvirt db.
This patch changes the restore flow such that it first removes extra
networks/bonds (which do not appear in the persisted config, but do
appear on the host) and only then attempts to restore the acquired
devices.
Change-Id: I387d5e3fd9be74e2789904ca8eaf9d867c9d4c3d
Signed-off-by: Edward Haas <edwardh(a)redhat.com>
---
M tests/functional/networkTests.py
M vdsm/vdsm-restore-net-config
2 files changed, 5 insertions(+), 5 deletions(-)
Approvals:
Jenkins CI: Passed CI tests
Petr Horáček: Looks good to me, but someone else must approve
Dan Kenigsberg: Looks good to me, approved
Edward Haas: Verified
--
To view, visit https://gerrit.ovirt.org/72324
To unsubscribe, visit https://gerrit.ovirt.org/settings
Gerrit-MessageType: merged
Gerrit-Change-Id: I387d5e3fd9be74e2789904ca8eaf9d867c9d4c3d
Gerrit-PatchSet: 5
Gerrit-Project: vdsm
Gerrit-Branch: master
Gerrit-Owner: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Dan Kenigsberg <danken(a)redhat.com>
Gerrit-Reviewer: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Jenkins CI
Gerrit-Reviewer: Petr Horáček <phoracek(a)redhat.com>
Gerrit-Reviewer: gerrit-hooks <automation(a)ovirt.org>
From Dan Kenigsberg <danken(a)redhat.com>:
Dan Kenigsberg has submitted this change and it was merged.
Change subject: net: Expose an existing bug in network restoration
......................................................................
net: Expose an existing bug in network restoration
This patch exposes a bug that exists in the network restoration flow
but was missed due to a missing step in one test and a missing
assertion.
testVolatileConfig test now simulates a reboot by deleting the running
config (which resides under /var/run/).
In addition, the assertNetworkDoesntExist assertion should also check
against the libvirt DB, assuring no leftovers are left there.
In the next patch this scenario will be solved, passing the failing
testVolatileConfig functional test.
Change-Id: Ifad0c04df9a36fba43ee678bbec4f3500b6a98ec
Signed-off-by: Edward Haas <edwardh(a)redhat.com>
---
M lib/vdsm/network/libvirt.py
M tests/functional/networkTests.py
2 files changed, 12 insertions(+), 0 deletions(-)
Approvals:
Jenkins CI: Passed CI tests
Dan Kenigsberg: Looks good to me, approved
Edward Haas: Verified
--
To view, visit https://gerrit.ovirt.org/72313
To unsubscribe, visit https://gerrit.ovirt.org/settings
Gerrit-MessageType: merged
Gerrit-Change-Id: Ifad0c04df9a36fba43ee678bbec4f3500b6a98ec
Gerrit-PatchSet: 4
Gerrit-Project: vdsm
Gerrit-Branch: master
Gerrit-Owner: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Dan Kenigsberg <danken(a)redhat.com>
Gerrit-Reviewer: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Jenkins CI
Gerrit-Reviewer: Petr Horáček <phoracek(a)redhat.com>
Gerrit-Reviewer: gerrit-hooks <automation(a)ovirt.org>
From Dan Kenigsberg <danken(a)redhat.com>:
Dan Kenigsberg has posted comments on this change.
Change subject: net: Modify restore-net to remove nets using setupNetworks
......................................................................
Patch Set 5: Code-Review+2
raising
--
To view, visit https://gerrit.ovirt.org/72324
To unsubscribe, visit https://gerrit.ovirt.org/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: I387d5e3fd9be74e2789904ca8eaf9d867c9d4c3d
Gerrit-PatchSet: 5
Gerrit-Project: vdsm
Gerrit-Branch: master
Gerrit-Owner: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Dan Kenigsberg <danken(a)redhat.com>
Gerrit-Reviewer: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Jenkins CI
Gerrit-Reviewer: Petr Horáček <phoracek(a)redhat.com>
Gerrit-Reviewer: gerrit-hooks <automation(a)ovirt.org>
Gerrit-HasComments: No
From Dan Kenigsberg <danken(a)redhat.com>:
Dan Kenigsberg has posted comments on this change.
Change subject: net: Expose an existing bug in network restoration
......................................................................
Patch Set 4: Code-Review+2
--
To view, visit https://gerrit.ovirt.org/72313
To unsubscribe, visit https://gerrit.ovirt.org/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: Ifad0c04df9a36fba43ee678bbec4f3500b6a98ec
Gerrit-PatchSet: 4
Gerrit-Project: vdsm
Gerrit-Branch: master
Gerrit-Owner: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Dan Kenigsberg <danken(a)redhat.com>
Gerrit-Reviewer: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Jenkins CI
Gerrit-Reviewer: Petr Horáček <phoracek(a)redhat.com>
Gerrit-Reviewer: gerrit-hooks <automation(a)ovirt.org>
Gerrit-HasComments: No
From Dan Kenigsberg <danken(a)redhat.com>:
Dan Kenigsberg has posted comments on this change.
Change subject: net tests: Wait for link aggregation bonds to be in sync
......................................................................
Patch Set 5:
(2 comments)
https://gerrit.ovirt.org/#/c/73042/5/tests/functional/networkTests.py
File tests/functional/networkTests.py:
Line 2791: ZEROMAC = '00:00:00:00:00:00'
Line 2792:
Line 2793: bond_1_partner_mac = ZEROMAC
Line 2794: bond_2_partner_mac = ZEROMAC
Line 2795: while bond_1_partner_mac == ZEROMAC or bond_2_partner_mac == ZEROMAC:
> add max attempts?
we must not block the test forever.
Line 2796: status, msg, info = self.vdsm_net.getVdsCapabilities()
Line 2797: bond_caps = info['bondings']
Line 2798: bond_1_partner_mac = \
Line 2799: bond_caps[BONDING_NAME]['ad_partner_mac']
Line 2818: self.assertEqual(status, SUCCESS, msg)
Line 2819: self.assertBondExists(bond, pair)
Line 2820:
Line 2821: if mode == LINK_AGGR_MODE:
Line 2822: self._wait_link_aggr_mode_sync()
so, haven't you found a nicer to wait for this sync, using netlink events?
_wait_link_aggr_mode_sync() would be slightly less ugly if it received the bond names as args.
Line 2823:
Line 2824: status, msg, info = self.vdsm_net.getVdsCapabilities()
Line 2825: bond_caps, nic_caps = info['bondings'], info['nics']
Line 2826: try:
--
To view, visit https://gerrit.ovirt.org/73042
To unsubscribe, visit https://gerrit.ovirt.org/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: I70ad2e8fbf2b54971eba79f8de44e199e54ae4e3
Gerrit-PatchSet: 5
Gerrit-Project: vdsm
Gerrit-Branch: master
Gerrit-Owner: Leon Goldberg <lgoldber(a)redhat.com>
Gerrit-Reviewer: Dan Kenigsberg <danken(a)redhat.com>
Gerrit-Reviewer: Edward Haas <edwardh(a)redhat.com>
Gerrit-Reviewer: Jenkins CI
Gerrit-Reviewer: Leon Goldberg <lgoldber(a)redhat.com>
Gerrit-Reviewer: gerrit-hooks <automation(a)ovirt.org>
Gerrit-HasComments: Yes
From Dan Kenigsberg <danken(a)redhat.com>:
Dan Kenigsberg has posted comments on this change.
Change subject: he_upgrade: add commands for 3.5 -> 3.6 upgrade with jsonrpc
......................................................................
Patch Set 2: Code-Review+1
The patch is so simple, that I'd take it in, even though the bug is not marked as an exception.
--
To view, visit https://gerrit.ovirt.org/72731
To unsubscribe, visit https://gerrit.ovirt.org/settings
Gerrit-MessageType: comment
Gerrit-Change-Id: Ie8c2575e5c5874fc4152fb1ea6f9eb3ff400fa9e
Gerrit-PatchSet: 2
Gerrit-Project: vdsm
Gerrit-Branch: ovirt-4.1
Gerrit-Owner: Simone Tiraboschi <stirabos(a)redhat.com>
Gerrit-Reviewer: Dan Kenigsberg <danken(a)redhat.com>
Gerrit-Reviewer: Francesco Romani <fromani(a)redhat.com>
Gerrit-Reviewer: Jenkins CI
Gerrit-Reviewer: Piotr Kliczewski <piotr.kliczewski(a)gmail.com>
Gerrit-Reviewer: Simone Tiraboschi <stirabos(a)redhat.com>
Gerrit-HasComments: No