| Summary: | Should disallow to use Negative values for 'max_*' in libvirtd.conf | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 7 | Reporter: | Yang Yang <yanyang> |
| Component: | libvirt | Assignee: | Michal Privoznik <mprivozn> |
| Status: | CLOSED ERRATA | QA Contact: | Virtualization Bugs <virt-bugs> |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | 7.1 | CC: | dyuan, lsu, mprivozn, mzhan, pkrempa, rbalakri, shyu, xuzhang |
| Target Milestone: | rc | Keywords: | Upstream |
| Target Release: | --- | ||
| Hardware: | x86_64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | libvirt-1.2.13-1.el7 | Doc Type: | Bug Fix |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2015-11-19 05:55:13 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
Hi Michal, As described in comment #8 in bug 992980, I opened this one to track the negative value issue. Thanks Yang Patches proposed upstream: https://www.redhat.com/archives/libvir-list/2014-December/msg00522.html And I've just pushed the patches upstream:
commit ca4f9518b8d381f8f14e37dc4d0367d33d5f3737
Author: Michal Privoznik <mprivozn>
AuthorDate: Tue Dec 9 16:22:09 2014 +0100
Commit: Michal Privoznik <mprivozn>
CommitDate: Mon Dec 15 10:34:18 2014 +0100
virconf: Introduce VIR_CONF_ULONG
https://bugzilla.redhat.com/show_bug.cgi?id=1160995
In our config files users are expected to pass several integer values
for different configuration knobs. However, majority of them expect a
nonnegative number and only a few of them accept a negative number too
(notably keepalive_interval in libvirtd.conf).
Therefore, a new type to config value is introduced: VIR_CONF_ULONG
that is set whenever an integer is positive or zero. With this
approach knobs accepting VIR_CONF_LONG should accept VIR_CONF_ULONG
too.
Signed-off-by: Michal Privoznik <mprivozn>
commit f81a70218081b68754280a6b71bbf48c2c1f4d6f
Author: Michal Privoznik <mprivozn>
AuthorDate: Tue Dec 9 14:53:28 2014 +0100
Commit: Michal Privoznik <mprivozn>
CommitDate: Mon Dec 15 10:34:18 2014 +0100
virConfType: switch to VIR_ENUM_{DECL,IMPL}
There's no need to implement ToString() function like we do if we
can use our shiny macros.
Signed-off-by: Michal Privoznik <mprivozn>
commit 4523b7769dc2baf91d63cd3cb2bf164e76825d78
Author: Michal Privoznik <mprivozn>
AuthorDate: Tue Dec 9 14:48:54 2014 +0100
Commit: Michal Privoznik <mprivozn>
CommitDate: Mon Dec 15 10:34:18 2014 +0100
virConfSetValue: Simplify condition
There's no need for condition of the following form:
if (str && STREQ(str, dst))
since we have STREQ_NULLABLE macro that handles NULL cases.
Signed-off-by: Michal Privoznik <mprivozn>
v1.2.11-9-gca4f951
I can produce this bug with build
libvirt-1.2.8-6.el7.x86_64
Verify it with build libvirt-1.2.15-2.el7.x86_64
1. set anyone of below value to negative value
#/etc/libvirt/libvirtd.conf
listen_tcp = -1
listen_tls = -40
mdns_adv = -1
tls_no_sanity_certificate = -1
max_clients = -1
max_queued_clients = -1
max_anonymous_clients = -1
max_workers = -1
min_workers = -1
prio_workers = -1
max_client_requests = -5
audit_level = -2
audit_logging = -1
log_level = -1
keepalive_count = -5
keepalive_required = -1
#/etc/libvirt/qemu.conf
remote_websocket_port_min = -5702
remote_websocket_port_max = -65535
remote_display_port_min = -5900
remote_display_port_max = -65535
migration_port_min = -49152
migration_port_max = -49215
max_processes = -5
max_files = -5
max_queued = -3
keepalive_count = -5
2. restart libvirtd service failed
# service libvirtd restart
Redirecting to /bin/systemctl restart libvirtd.service
Job for libvirtd.service failed. See 'systemctl status libvirtd.service' and 'journalctl -xn' for details.
[root@server libvirt]# service libvirtd status
Redirecting to /bin/systemctl status libvirtd.service
libvirtd.service - Virtualization daemon
Loaded: loaded (/usr/lib/systemd/system/libvirtd.service; disabled)
Active: failed (Result: start-limit) since Mon 2015-06-01 12:56:44 HKT; 2s ago
Docs: man:libvirtd(8)
http://libvirt.org
Process: 29089 ExecStart=/usr/sbin/libvirtd $LIBVIRTD_ARGS (code=exited, status=1/FAILURE)
Main PID: 29089 (code=exited, status=1/FAILURE)
Jun 01 12:56:44 server.englab.nay.redhat.com systemd[1]: Failed to start Virtualization daemon.
Jun 01 12:56:44 server.englab.nay.redhat.com systemd[1]: Unit libvirtd.service entered failed state.
Jun 01 12:56:44 server.englab.nay.redhat.com systemd[1]: libvirtd.service holdoff time over, scheduling restart.
Jun 01 12:56:44 server.englab.nay.redhat.com systemd[1]: Stopping Virtualization daemon...
Jun 01 12:56:44 server.englab.nay.redhat.com systemd[1]: Starting Virtualization daemon...
Jun 01 12:56:44 server.englab.nay.redhat.com systemd[1]: libvirtd.service start request repeated too quickly, refusing to start.
Jun 01 12:56:44 server.englab.nay.redhat.com systemd[1]: Failed to start Virtualization daemon.
Jun 01 12:56:44 server.englab.nay.redhat.com systemd[1]: Unit libvirtd.service entered failed state.
#cat /var/log/messages, find below error messages
...
Jun 1 12:56:44 server libvirtd: 2015-06-01 04:56:44.775+0000: 29089: error : main:1234 : Can't load config file: unsupported configuration: remoteReadConfigFile: /etc/libvirt/libvirtd.conf: max_clients: invalid type: got long; expected unsigned long: /etc/libvirt/libvirtd.conf
Jun 1 12:56:44 server systemd: libvirtd.service: main process exited, code=exited, status=1/FAILURE
Jun 1 12:56:44 server systemd: Failed to start Virtualization daemon.
Jun 1 12:56:44 server systemd: Unit libvirtd.service entered failed state.
Jun 1 12:56:44 server systemd: libvirtd.service holdoff time over, scheduling restart.
Jun 1 12:56:44 server systemd: Stopping Virtualization daemon...
Jun 1 12:56:44 server systemd: Starting Virtualization daemon...
Jun 1 12:56:44 server systemd: libvirtd.service start request repeated too quickly, refusing to start.
Jun 1 12:56:44 server systemd: Failed to start Virtualization daemon.
Jun 1 12:56:44 server systemd: Unit libvirtd.service entered failed state.
~
3. set anyone of them to positive value or zero, libvirtd service start success
4. when configure max_requests and keepalive_interval to negative value, such as -1, -5, -20, libvirtd service start success
check patch, these two parameters missing to modify, is this expected result?
#git show ca4f9518b8d381f8f14e37dc4d0367d33d5f3737
- GET_CONF_INT(conf, filename, min_workers);
- GET_CONF_INT(conf, filename, max_workers);
- GET_CONF_INT(conf, filename, max_clients);
- GET_CONF_INT(conf, filename, max_queued_clients);
- GET_CONF_INT(conf, filename, max_anonymous_clients);
+ GET_CONF_UINT(conf, filename, min_workers);
+ GET_CONF_UINT(conf, filename, max_workers);
+ GET_CONF_UINT(conf, filename, max_clients);
+ GET_CONF_UINT(conf, filename, max_queued_clients);
+ GET_CONF_UINT(conf, filename, max_anonymous_clients);
- GET_CONF_INT(conf, filename, prio_workers);
+ GET_CONF_UINT(conf, filename, prio_workers);
GET_CONF_INT(conf, filename, max_requests); <---- notice this line
- GET_CONF_INT(conf, filename, max_client_requests);
+ GET_CONF_UINT(conf, filename, max_client_requests);
GET_CONF_INT(conf, filename, keepalive_interval); <---- notice this line
- GET_CONF_INT(conf, filename, keepalive_count);
- GET_CONF_INT(conf, filename, keepalive_required);
+ GET_CONF_UINT(conf, filename, keepalive_count);
+ GET_CONF_UINT(conf, filename, keepalive_required);
5. # grep lock_ /etc/libvirt/qemu.conf
lock_manager = "lockd"
6.set anyone of below parameter to negative value in
#/etc/libvirt/qemu-lockd.conf
auto_disk_leases = -2
require_lease_for_disks = -10
# service libvirtd restart
Redirecting to /bin/systemctl restart libvirtd.service
[root@server libvirt]# service libvirtd status
Redirecting to /bin/systemctl status libvirtd.service
libvirtd.service - Virtualization daemon
Loaded: loaded (/usr/lib/systemd/system/libvirtd.service; disabled)
Active: inactive (dead) since Mon 2015-06-01 17:32:44 HKT; 897ms ago
Docs: man:libvirtd(8)
http://libvirt.org
Process: 6959 ExecStart=/usr/sbin/libvirtd $LIBVIRTD_ARGS (code=exited, status=0/SUCCESS)
Main PID: 6959 (code=exited, status=0/SUCCESS)
Jun 01 17:32:44 server.englab.nay.redhat.com libvirtd[6959]: 2015-06-01 09:32:44.164+0000: 7035: debug : virFileCl... 13
Jun 01 17:32:44 server.englab.nay.redhat.com libvirtd[6959]: 2015-06-01 09:32:44.164+0000: 7035: debug : virFileCl... 14
Jun 01 17:32:44 server.englab.nay.redhat.com libvirtd[6959]: 2015-06-01 09:32:44.164+0000: 7035: debug : virFileCl... 15
Jun 01 17:32:44 server.englab.nay.redhat.com libvirtd[6959]: 2015-06-01 09:32:44.164+0000: 7035: debug : virFileCl... 16
Jun 01 17:32:44 server.englab.nay.redhat.com libvirtd[6959]: 2015-06-01 09:32:44.164+0000: 7035: debug : virFileCl... 18
Jun 01 17:32:44 server.englab.nay.redhat.com libvirtd[6959]: 2015-06-01 09:32:44.164+0000: 7035: debug : virFileCl... 20
Jun 01 17:32:44 server.englab.nay.redhat.com libvirtd[6959]: 2015-06-01 09:32:44.166+0000: 7036: debug : virFileCl...d 3
Jun 01 17:32:44 server.englab.nay.redhat.com libvirtd[6959]: 2015-06-01 09:32:44.166+0000: 7036: debug : virFileCl...d 5
Jun 01 17:32:44 server.englab.nay.redhat.com libvirtd[6959]: 2015-06-01 09:32:44.166+0000: 7036: debug : virFileCl...d 6
Jun 01 17:32:44 server.englab.nay.redhat.com libvirtd[6959]: 2015-06-01 09:32:44.166+0000: 7036: debug : virFileCl...d 7
Hint: Some lines were ellipsized, use -l to show in full.
7. set anyone of below value
#/etc/libvirt/virtlockd.conf
log_level = -3
max_clients = -1024
service virtlockd start failed
systemctl status virtlockd.service
virtlockd.service - Virtual machine lock manager
Loaded: loaded (/usr/lib/systemd/system/virtlockd.service; static)
Active: failed (Result: exit-code) since Mon 2015-06-01 17:27:15 HKT; 5s ago
Docs: man:virtlockd(8)
http://libvirt.org
Process: 6487 ExecStart=/usr/sbin/virtlockd $VIRTLOCKD_ARGS (code=exited, status=1/FAILURE)
Main PID: 6487 (code=exited, status=1/FAILURE)
Jun 01 17:27:14 server.englab.nay.redhat.com systemd[1]: Started Virtual machine lock manager.
Jun 01 17:27:15 server.englab.nay.redhat.com virtlockd[6487]: 2015-06-01 09:27:15.053+0000: 6487: info : libvirt v...om)
Jun 01 17:27:15 server.englab.nay.redhat.com virtlockd[6487]: 2015-06-01 09:27:15.053+0000: 6487: error : main:123...onf
Jun 01 17:27:15 server.englab.nay.redhat.com systemd[1]: virtlockd.service: main process exited, code=exited, sta...LURE
Jun 01 17:27:15 server.englab.nay.redhat.com systemd[1]: Unit virtlockd.service entered failed state.
Hint: Some lines were ellipsized, use -l to show in full.
please help me confirm whether step 4 result are expected? and are the above steps enough to verify this bug?
thanks
(In reply to vivian zhang from comment #7) > 4. when configure max_requests and keepalive_interval to negative value, > such as -1, -5, -20, libvirtd service start success > > check patch, these two parameters missing to modify, is this expected result? > > #git show ca4f9518b8d381f8f14e37dc4d0367d33d5f3737 > > GET_CONF_INT(conf, filename, max_requests); <---- notice this line > GET_CONF_INT(conf, filename, keepalive_interval); <---- notice this > line This is correct. As it says in libvirtd.conf, max_requests is not used anywhere, not enforced. keepalive_interval can be a negative value in which case it means that keepalive is disabled. > please help me confirm whether step 4 result are expected? and are the above > steps enough to verify this bug? Yes, everything seems to be working. > > thanks You are welcome. according to commet8, set this bug to verified Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://rhn.redhat.com/errata/RHBA-2015-2202.html |
Description of problem: Set negative value to 'max_*' (e.g. max_clients, max_queued_clients, max_anonymous_clients, max_workers), it is translated to a large positive value. Libvirt should disallow negative value to be used for 'max_*' Version-Release number of selected component (if applicable): libvirt-1.2.8-6.el7.x86_64 How reproducible: 100% Steps to Reproduce: 1. Set 'max_*' as following # grep max /etc/libvirt/libvirtd.conf max_clients= -40 max_queued_clients = 0 max_anonymous_clients = -1 max_workers = -40 2. start 50 lxc container # for i in {1..50}; do virsh -c lxc:/// start lxc-test-$i & done 3. # grep virNetServerCheckLimits /var/log/libvirtd.log 2014-11-06 05:20:36.470+0000: 16574: debug : virNetServerCheckLimits:1078 : Considering re-enabling services: nclients=5 nclients_max=18446744073709551576 nclients_unauth=4 nclients_unauth_max=18446744073709551615 2014-11-06 05:20:36.470+0000: 16574: debug : virNetServerCheckLimits:1083 : Re-enabling services Actual results: Setting '-40' to 'max_clients', it is translated into 18446744073709551576 Setting '-1' to 'max_anonymous_clients', it is translated into 18446744073709551615 Expected results: disable negative value Additional info: