I found that it is getting checked in the first fit allocator
(server/src/com/cloud/agent/manager/allocator/impl/FirstFitAllocator.java). The
'==' check is incorrect though, should have been '>='. Please go ahead and file
a bug if this is not working as expected.
Code snippet
//find number of guest VMs occupying capacity on this host.
Long vmCount = _vmInstanceDao.countRunningByHostId(host.getId());
Long maxGuestLimit = getHostMaxGuestLimit(host);
if (vmCount.longValue() == maxGuestLimit.longValue()){
if (s_logger.isDebugEnabled()) {
s_logger.debug("Host name: " + host.getName() + ", hostId:
"+ host.getId() +" already has max Running VMs(count includes system VMs),
limit is: " + maxGuestLimit + " , skipping this and trying other available
hosts");
}
continue;
}
-----Original Message-----
From: Mice Xia [mailto:[email protected]]
Sent: Tuesday, July 24, 2012 11:33 AM
To: [email protected]; [email protected]
Cc: Mice Xia
Subject: 答复: max guest limit in hypervisor capabilities
Hi, Koushik
SELECT * FROM cloud.hypervisor_capabilities;
+----+--------------------------------------+-----------------+--------------------+------------------+------------------------+
| id | uuid | hypervisor_type |
hypervisor_version | max_guests_limit | security_group_enabled |
+----+--------------------------------------+-----------------+--------------------+------------------+------------------------+
| 1 | fb09cd73-d2ff-4f11-84bb-ea0ae7d358ee | XenServer | default
| 2 | 1 |
| 2 | 518962ab-f50b-4e95-aaf2-aaece3e6c892 | XenServer | XCP 1.0
| 50 | 1 |
| 3 | e7feccc6-415b-440d-bc9d-fd9ffbfa3c47 | XenServer | 5.6
| 50 | 1 |
| 4 | 8107d8e2-b5dd-4041-a532-dc996e2d012a | XenServer | 5.6 FP1
| 50 | 1 |
| 5 | 50aa2194-e747-4667-8c75-0e0203826b5a | XenServer | 5.6 SP2
| 2 | 1 |
As it is, limits of Xenserver default and xenserver 5.6 sp2 were already
modified to 2 Not like a UI related issue.
Regards
Mice
