You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@cloudstack.apache.org by Marlon Davids <md...@umbee.co.uk> on 2012/04/25 11:04:09 UTC
Cloudstack Questions
Hello Community,
We have been working with Cloudstack for a while now and I thought it would be about time we participated in the community.
We have a couple of outstanding questions that I would really like if someone could answer:
1) Will NFS be phased out for secondary storage in the future?
2) How do we monitor VM's that are in Cloudstack when they are in an isolated VLAN does anyone have a clever workaround?
3) Has anyone developed a script for parsing and alerting on warning events in the management Log yet?
Kind Regards,
Marlon Davids
..............................................................................................
P: +44 (0)845 619 7444 | M: +44 (0)7956 920 496 E: mdavids@umbee.co.uk<ma...@umbee.co.uk>
..............................................................................................
W: www.umbee.co.uk<http://www.umbee.co.uk/> Managed Hosting Specialists.
..............................................................................................
Umbee Ltd, TMS House, Cray Avenue, Orpington, Kent, BR5 3QB
Company Registration Number. 06954604. VAT Registration Number 977446172
This e-mail and the information it contains are confidential.
If you have received this e-mail in error please notify the sender immediately.
You should not copy it for any purpose or disclose its contents to any other person.
The contents of this e-mail do not necessarily reflect the views of the company.
E&OE
P Please consider the environment before printing this email
Re: Cloudstack Questions
Posted by Chiradeep Vittal <Ch...@citrix.com>.
I put it on the Community page
http://goo.gl/4V94A
On 4/26/12 11:57 AM, "Marlon Davids" <md...@umbee.co.uk> wrote:
>Brilliant, thanks Chiradeep,
>
>
>
>Unfortunately I don't see the attachment?
>
>
>
>>There's a few approaches to the "monitoring VMs" question.
>
>A. Install agents on the VMs and get the agents to phone back to your
>monitoring server.
>
>B. Add a dedicated monitoring VLAN per tenant and get the tenant VMs to
>be dual homed on the original isolated VLAN and the monitoring VLAN C.
>Like B, but the VLAN is shared among tenants. There is no isolation
>guarantees in this case.
>
>D. Create a monitoring VLAN. Every tenant creates (or admin creates for
>
>them) a new routing VM (can be any old Linux distro) that has an
>interface on both the monitoring VLAN and the tenant's VLAN. Tenant VMs
>add static routes to the monitoring server to go via the new routing VM.
>
>
>
>See attached PDF
>
>You can use (D) to add any special networking service. For example you
>can add a site-to-site VPN using the additional routing VM (the
>monitoring VLAN becomes a VLAN with public Ips).
>
>
>Kind Regards,
>Marlon Davids
>
>..........................................................................
>....................
>P: +44 (0)845 619 7444 | E:
>mdavids@umbee.co.uk<ma...@umbee.co.uk>
>..........................................................................
>....................
>W: www.umbee.co.uk<http://www.umbee.co.uk/> Managed Hosting Specialists.
>..........................................................................
>....................
>Umbee Ltd, TMS House, Cray Avenue, Orpington, Kent, BR5 3QB
>Company Registration Number. 06954604. VAT Registration Number 977446172
>
>This e-mail and the information it contains are confidential.
>If you have received this e-mail in error please notify the sender
>immediately.
>You should not copy it for any purpose or disclose its contents to any
>other person.
>The contents of this e-mail do not necessarily reflect the views of the
>company.
>E&OE
>
>P Please consider the environment before printing this email
>
>
>
>-----Original Message-----
>From: Chiradeep Vittal [mailto:Chiradeep.Vittal@citrix.com]
>Sent: 26 April 2012 18:19
>To: cloudstack-dev@incubator.apache.org
>Subject: Re: Cloudstack Questions
>
>
>
>There's a few approaches to the "monitoring VMs" question.
>
>A. Install agents on the VMs and get the agents to phone back to your
>monitoring server.
>
>B. Add a dedicated monitoring VLAN per tenant and get the tenant VMs to
>be dual homed on the original isolated VLAN and the monitoring VLAN C.
>Like B, but the VLAN is shared among tenants. There is no isolation
>guarantees in this case.
>
>D. Create a monitoring VLAN. Every tenant creates (or admin creates for
>
>them) a new routing VM (can be any old Linux distro) that has an
>interface on both the monitoring VLAN and the tenant's VLAN. Tenant VMs
>add static routes to the monitoring server to go via the new routing VM.
>
>
>
>See attached PDF
>
>You can use (D) to add any special networking service. For example you
>can add a site-to-site VPN using the additional routing VM (the
>monitoring VLAN becomes a VLAN with public Ips).
>
>
>
>
>
>On 4/25/12 4:49 AM, "David Nalley" <da...@gnsa.us>>
>wrote:
>
>
>
>>Welcome Marlon
>
>>
>
>>>
>
>>> 1) Will NFS be phased out for secondary storage in the future?
>
>>
>
>>I don't think this is going to happen - and if it is, certainly not any
>
>>time soon.
>
>>
>
>>> 2) How do we monitor VM's that are in Cloudstack when they are in an
>
>>>isolated VLAN does anyone have a clever workaround?
>
>>
>
>>I dont have a good answer for this at the moment
>
>>
>
>>> 3) Has anyone developed a script for parsing and alerting on warning
>
>>>events in the management Log yet?
>
>>>
>
>>
>
>>So you should check out Zenoss's CloudStack ZenPack [1] they poll the
>
>>listEvents API call and parse that for events.
>
>>Alternatively, you could configure log4j to send logs of warning or
>
>>higher to syslog - and thus on to some central logging facility like
>
>>logstash, greylog, or splunk.
>
>>
>
>>
>
>>
>
>>[1] https://github.com/zenoss/ZenPacks.zenoss.CloudStack
>
>
RE: Cloudstack Questions
Posted by Marlon Davids <md...@umbee.co.uk>.
Brilliant, thanks Chiradeep,
Unfortunately I don't see the attachment?
>There's a few approaches to the "monitoring VMs" question.
A. Install agents on the VMs and get the agents to phone back to your monitoring server.
B. Add a dedicated monitoring VLAN per tenant and get the tenant VMs to be dual homed on the original isolated VLAN and the monitoring VLAN C. Like B, but the VLAN is shared among tenants. There is no isolation guarantees in this case.
D. Create a monitoring VLAN. Every tenant creates (or admin creates for
them) a new routing VM (can be any old Linux distro) that has an interface on both the monitoring VLAN and the tenant's VLAN. Tenant VMs add static routes to the monitoring server to go via the new routing VM.
See attached PDF
You can use (D) to add any special networking service. For example you can add a site-to-site VPN using the additional routing VM (the monitoring VLAN becomes a VLAN with public Ips).
Kind Regards,
Marlon Davids
..............................................................................................
P: +44 (0)845 619 7444 | E: mdavids@umbee.co.uk<ma...@umbee.co.uk>
..............................................................................................
W: www.umbee.co.uk<http://www.umbee.co.uk/> Managed Hosting Specialists.
..............................................................................................
Umbee Ltd, TMS House, Cray Avenue, Orpington, Kent, BR5 3QB
Company Registration Number. 06954604. VAT Registration Number 977446172
This e-mail and the information it contains are confidential.
If you have received this e-mail in error please notify the sender immediately.
You should not copy it for any purpose or disclose its contents to any other person.
The contents of this e-mail do not necessarily reflect the views of the company.
E&OE
P Please consider the environment before printing this email
-----Original Message-----
From: Chiradeep Vittal [mailto:Chiradeep.Vittal@citrix.com]
Sent: 26 April 2012 18:19
To: cloudstack-dev@incubator.apache.org
Subject: Re: Cloudstack Questions
There's a few approaches to the "monitoring VMs" question.
A. Install agents on the VMs and get the agents to phone back to your monitoring server.
B. Add a dedicated monitoring VLAN per tenant and get the tenant VMs to be dual homed on the original isolated VLAN and the monitoring VLAN C. Like B, but the VLAN is shared among tenants. There is no isolation guarantees in this case.
D. Create a monitoring VLAN. Every tenant creates (or admin creates for
them) a new routing VM (can be any old Linux distro) that has an interface on both the monitoring VLAN and the tenant's VLAN. Tenant VMs add static routes to the monitoring server to go via the new routing VM.
See attached PDF
You can use (D) to add any special networking service. For example you can add a site-to-site VPN using the additional routing VM (the monitoring VLAN becomes a VLAN with public Ips).
On 4/25/12 4:49 AM, "David Nalley" <da...@gnsa.us>> wrote:
>Welcome Marlon
>
>>
>> 1) Will NFS be phased out for secondary storage in the future?
>
>I don't think this is going to happen - and if it is, certainly not any
>time soon.
>
>> 2) How do we monitor VM's that are in Cloudstack when they are in an
>>isolated VLAN does anyone have a clever workaround?
>
>I dont have a good answer for this at the moment
>
>> 3) Has anyone developed a script for parsing and alerting on warning
>>events in the management Log yet?
>>
>
>So you should check out Zenoss's CloudStack ZenPack [1] they poll the
>listEvents API call and parse that for events.
>Alternatively, you could configure log4j to send logs of warning or
>higher to syslog - and thus on to some central logging facility like
>logstash, greylog, or splunk.
>
>
>
>[1] https://github.com/zenoss/ZenPacks.zenoss.CloudStack
RE: Cloudstack Questions
Posted by Clayton Weise <cw...@iswest.net>.
We haven't had a chance to kick the tires on it but we came across a commercial product called PHD Virtual which is a XenServer specific monitoring application. It apparently has the ability to grab the type of data you're looking for directly from XAPI. Zenoss might have an option for this as well, and I know that Zenoss has a CloudStack plugin as well for pulling data from your management server(s). We didn't get a definitive answer out of the Zenoss people about getting details on guests but we haven't had a chance to explore it much either.
-----Original Message-----
From: Marlon Davids [mailto:mdavids@umbee.co.uk]
Sent: Wednesday, April 25, 2012 9:52 AM
To: cloudstack-dev@incubator.apache.org
Subject: RE: Cloudstack Questions
> It depends on two things: What you're trying to monitor and what hypervisor you have. We've been looking into this subject and we want to poll basic monitoring data (cpu, memory, disk, etc) which can be done at the hypervisor level with the right monitoring tool.
We use Xenserver and want to do basic Ping checks and cpu, memory, disk, ?
We have Nagios in place so I will look to see what level of hypervisor monitoring can be achieved.
Kind Regards,
Marlon Davids
..............................................................................................
P: +44 (0)845 619 7444 | M: +44 (0)7827 333 583 E: mdavids@umbee.co.uk
..............................................................................................
W: www.umbee.co.uk Managed Hosting Specialists.
..............................................................................................
Umbee Ltd, TMS House, Cray Avenue, Orpington, Kent, BR5 3QB
Company Registration Number. 06954604. VAT Registration Number 977446172
This e-mail and the information it contains are confidential.
If you have received this e-mail in error please notify the sender immediately.
You should not copy it for any purpose or disclose its contents to any other person.
The contents of this e-mail do not necessarily reflect the views of the company.
E&OE
Please consider the environment before printing this email
-----Original Message-----
From: Clayton Weise [mailto:cweise@iswest.net]
Sent: 25 April 2012 17:01
To: 'cloudstack-dev@incubator.apache.org'
Subject: RE: Cloudstack Questions
> 2) How do we monitor VM's that are in Cloudstack when they are in an isolated VLAN does anyone have a clever workaround?
It depends on two things: What you're trying to monitor and what hypervisor you have. We've been looking into this subject and we want to poll basic monitoring data (cpu, memory, disk, etc) which can be done at the hypervisor level with the right monitoring tool.
To get beyond that and dig down into the guests the only thing we've come up with is by installing/configuring a monitoring agent on all of our templates so that when the template is deployed it will automatically update our monitor with info about the guest VM. I guess you could also look at modifying the virtual router system VM and deploying something on there that would discover all of the VMs, pull from them, and send it back to a central monitoring system. But that adds more work for what is already a pretty light weight installation so it might not be desirable.
-Clayton
-----Original Message-----
From: David Nalley [mailto:david@gnsa.us]
Sent: Wednesday, April 25, 2012 4:50 AM
To: cloudstack-dev@incubator.apache.org
Subject: Re: Cloudstack Questions
Welcome Marlon
>
> 1) Will NFS be phased out for secondary storage in the future?
I don't think this is going to happen - and if it is, certainly not any time soon.
> 2) How do we monitor VM's that are in Cloudstack when they are in an isolated VLAN does anyone have a clever workaround?
I dont have a good answer for this at the moment
> 3) Has anyone developed a script for parsing and alerting on warning events in the management Log yet?
>
So you should check out Zenoss's CloudStack ZenPack [1] they poll the listEvents API call and parse that for events.
Alternatively, you could configure log4j to send logs of warning or higher to syslog - and thus on to some central logging facility like logstash, greylog, or splunk.
[1] https://github.com/zenoss/ZenPacks.zenoss.CloudStack
RE: Cloudstack Questions
Posted by Marlon Davids <md...@umbee.co.uk>.
> It depends on two things: What you're trying to monitor and what hypervisor you have. We've been looking into this subject and we want to poll basic monitoring data (cpu, memory, disk, etc) which can be done at the hypervisor level with the right monitoring tool.
We use Xenserver and want to do basic Ping checks and cpu, memory, disk, ?
We have Nagios in place so I will look to see what level of hypervisor monitoring can be achieved.
Kind Regards,
Marlon Davids
..............................................................................................
P: +44 (0)845 619 7444 | M: +44 (0)7827 333 583 E: mdavids@umbee.co.uk
..............................................................................................
W: www.umbee.co.uk Managed Hosting Specialists.
..............................................................................................
Umbee Ltd, TMS House, Cray Avenue, Orpington, Kent, BR5 3QB
Company Registration Number. 06954604. VAT Registration Number 977446172
This e-mail and the information it contains are confidential.
If you have received this e-mail in error please notify the sender immediately.
You should not copy it for any purpose or disclose its contents to any other person.
The contents of this e-mail do not necessarily reflect the views of the company.
E&OE
Please consider the environment before printing this email
-----Original Message-----
From: Clayton Weise [mailto:cweise@iswest.net]
Sent: 25 April 2012 17:01
To: 'cloudstack-dev@incubator.apache.org'
Subject: RE: Cloudstack Questions
> 2) How do we monitor VM's that are in Cloudstack when they are in an isolated VLAN does anyone have a clever workaround?
It depends on two things: What you're trying to monitor and what hypervisor you have. We've been looking into this subject and we want to poll basic monitoring data (cpu, memory, disk, etc) which can be done at the hypervisor level with the right monitoring tool.
To get beyond that and dig down into the guests the only thing we've come up with is by installing/configuring a monitoring agent on all of our templates so that when the template is deployed it will automatically update our monitor with info about the guest VM. I guess you could also look at modifying the virtual router system VM and deploying something on there that would discover all of the VMs, pull from them, and send it back to a central monitoring system. But that adds more work for what is already a pretty light weight installation so it might not be desirable.
-Clayton
-----Original Message-----
From: David Nalley [mailto:david@gnsa.us]
Sent: Wednesday, April 25, 2012 4:50 AM
To: cloudstack-dev@incubator.apache.org
Subject: Re: Cloudstack Questions
Welcome Marlon
>
> 1) Will NFS be phased out for secondary storage in the future?
I don't think this is going to happen - and if it is, certainly not any time soon.
> 2) How do we monitor VM's that are in Cloudstack when they are in an isolated VLAN does anyone have a clever workaround?
I dont have a good answer for this at the moment
> 3) Has anyone developed a script for parsing and alerting on warning events in the management Log yet?
>
So you should check out Zenoss's CloudStack ZenPack [1] they poll the listEvents API call and parse that for events.
Alternatively, you could configure log4j to send logs of warning or higher to syslog - and thus on to some central logging facility like logstash, greylog, or splunk.
[1] https://github.com/zenoss/ZenPacks.zenoss.CloudStack
RE: Cloudstack Questions
Posted by Clayton Weise <cw...@iswest.net>.
> 2) How do we monitor VM's that are in Cloudstack when they are in an isolated VLAN does anyone have a clever workaround?
It depends on two things: What you're trying to monitor and what hypervisor you have. We've been looking into this subject and we want to poll basic monitoring data (cpu, memory, disk, etc) which can be done at the hypervisor level with the right monitoring tool.
To get beyond that and dig down into the guests the only thing we've come up with is by installing/configuring a monitoring agent on all of our templates so that when the template is deployed it will automatically update our monitor with info about the guest VM. I guess you could also look at modifying the virtual router system VM and deploying something on there that would discover all of the VMs, pull from them, and send it back to a central monitoring system. But that adds more work for what is already a pretty light weight installation so it might not be desirable.
-Clayton
-----Original Message-----
From: David Nalley [mailto:david@gnsa.us]
Sent: Wednesday, April 25, 2012 4:50 AM
To: cloudstack-dev@incubator.apache.org
Subject: Re: Cloudstack Questions
Welcome Marlon
>
> 1) Will NFS be phased out for secondary storage in the future?
I don't think this is going to happen - and if it is, certainly not
any time soon.
> 2) How do we monitor VM's that are in Cloudstack when they are in an isolated VLAN does anyone have a clever workaround?
I dont have a good answer for this at the moment
> 3) Has anyone developed a script for parsing and alerting on warning events in the management Log yet?
>
So you should check out Zenoss's CloudStack ZenPack [1] they poll the
listEvents API call and parse that for events.
Alternatively, you could configure log4j to send logs of warning or
higher to syslog - and thus on to some central logging facility like
logstash, greylog, or splunk.
[1] https://github.com/zenoss/ZenPacks.zenoss.CloudStack
Re: Cloudstack Questions
Posted by John Kinsella <jl...@stratosec.co>.
I was just about to see if anybody wanted to work on a nagios plugin, but googled first - https://github.com/jasonhancock/nagios-cloudstack looks interesting, for any other Opsview/Nagios peeps out there.
MIght be good to have a wiki page pointing to plugins for the various monitoring solutions.
John
On Apr 26, 2012, at 10:19 AM, Chiradeep Vittal wrote:
> There's a few approaches to the "monitoring VMs" question.
> A. Install agents on the VMs and get the agents to phone back to your
> monitoring server.
> B. Add a dedicated monitoring VLAN per tenant and get the tenant VMs to be
> dual homed on the original isolated VLAN and the monitoring VLAN
> C. Like B, but the VLAN is shared among tenants. There is no isolation
> guarantees in this case.
> D. Create a monitoring VLAN. Every tenant creates (or admin creates for
> them) a new routing VM (can be any old Linux distro) that has an interface
> on both the monitoring VLAN and the tenant's VLAN. Tenant VMs add static
> routes to the monitoring server to go via the new routing VM.
>
> See attached PDF
> You can use (D) to add any special networking service. For example you can
> add a site-to-site VPN using the additional routing VM (the monitoring
> VLAN becomes a VLAN with public Ips).
>
>
> On 4/25/12 4:49 AM, "David Nalley" <da...@gnsa.us> wrote:
>
>> Welcome Marlon
>>
>>>
>>> 1) Will NFS be phased out for secondary storage in the future?
>>
>> I don't think this is going to happen - and if it is, certainly not
>> any time soon.
>>
>>> 2) How do we monitor VM's that are in Cloudstack when they are in an
>>> isolated VLAN does anyone have a clever workaround?
>>
>> I dont have a good answer for this at the moment
>>
>>> 3) Has anyone developed a script for parsing and alerting on warning
>>> events in the management Log yet?
>>>
>>
>> So you should check out Zenoss's CloudStack ZenPack [1] they poll the
>> listEvents API call and parse that for events.
>> Alternatively, you could configure log4j to send logs of warning or
>> higher to syslog - and thus on to some central logging facility like
>> logstash, greylog, or splunk.
>>
>>
>>
>> [1] https://github.com/zenoss/ZenPacks.zenoss.CloudStack
>
Stratosec - Secure Infrastructure as a Service
o: 415.315.9385
@johnlkinsella
Re: Cloudstack Questions
Posted by David Noland <Da...@citrix.com>.
Is monitoring VM's through hypervisor APIs an option as well?
-David
On 4/26/12 10:19 AM, "Chiradeep Vittal" <Ch...@citrix.com>
wrote:
>There's a few approaches to the "monitoring VMs" question.
>A. Install agents on the VMs and get the agents to phone back to your
>monitoring server.
>B. Add a dedicated monitoring VLAN per tenant and get the tenant VMs to be
>dual homed on the original isolated VLAN and the monitoring VLAN
>C. Like B, but the VLAN is shared among tenants. There is no isolation
>guarantees in this case.
>D. Create a monitoring VLAN. Every tenant creates (or admin creates for
>them) a new routing VM (can be any old Linux distro) that has an interface
>on both the monitoring VLAN and the tenant's VLAN. Tenant VMs add static
>routes to the monitoring server to go via the new routing VM.
>
>See attached PDF
>You can use (D) to add any special networking service. For example you can
>add a site-to-site VPN using the additional routing VM (the monitoring
>VLAN becomes a VLAN with public Ips).
>
>
>On 4/25/12 4:49 AM, "David Nalley" <da...@gnsa.us> wrote:
>
>>Welcome Marlon
>>
>>>
>>> 1) Will NFS be phased out for secondary storage in the future?
>>
>>I don't think this is going to happen - and if it is, certainly not
>>any time soon.
>>
>>> 2) How do we monitor VM's that are in Cloudstack when they are in an
>>>isolated VLAN does anyone have a clever workaround?
>>
>>I dont have a good answer for this at the moment
>>
>>> 3) Has anyone developed a script for parsing and alerting on warning
>>>events in the management Log yet?
>>>
>>
>>So you should check out Zenoss's CloudStack ZenPack [1] they poll the
>>listEvents API call and parse that for events.
>>Alternatively, you could configure log4j to send logs of warning or
>>higher to syslog - and thus on to some central logging facility like
>>logstash, greylog, or splunk.
>>
>>
>>
>>[1] https://github.com/zenoss/ZenPacks.zenoss.CloudStack
>
Re: Cloudstack Questions
Posted by Chiradeep Vittal <Ch...@citrix.com>.
There's a few approaches to the "monitoring VMs" question.
A. Install agents on the VMs and get the agents to phone back to your
monitoring server.
B. Add a dedicated monitoring VLAN per tenant and get the tenant VMs to be
dual homed on the original isolated VLAN and the monitoring VLAN
C. Like B, but the VLAN is shared among tenants. There is no isolation
guarantees in this case.
D. Create a monitoring VLAN. Every tenant creates (or admin creates for
them) a new routing VM (can be any old Linux distro) that has an interface
on both the monitoring VLAN and the tenant's VLAN. Tenant VMs add static
routes to the monitoring server to go via the new routing VM.
See attached PDF
You can use (D) to add any special networking service. For example you can
add a site-to-site VPN using the additional routing VM (the monitoring
VLAN becomes a VLAN with public Ips).
On 4/25/12 4:49 AM, "David Nalley" <da...@gnsa.us> wrote:
>Welcome Marlon
>
>>
>> 1) Will NFS be phased out for secondary storage in the future?
>
>I don't think this is going to happen - and if it is, certainly not
>any time soon.
>
>> 2) How do we monitor VM's that are in Cloudstack when they are in an
>>isolated VLAN does anyone have a clever workaround?
>
>I dont have a good answer for this at the moment
>
>> 3) Has anyone developed a script for parsing and alerting on warning
>>events in the management Log yet?
>>
>
>So you should check out Zenoss's CloudStack ZenPack [1] they poll the
>listEvents API call and parse that for events.
>Alternatively, you could configure log4j to send logs of warning or
>higher to syslog - and thus on to some central logging facility like
>logstash, greylog, or splunk.
>
>
>
>[1] https://github.com/zenoss/ZenPacks.zenoss.CloudStack
Re: Cloudstack Questions
Posted by David Nalley <da...@gnsa.us>.
Welcome Marlon
>
> 1) Will NFS be phased out for secondary storage in the future?
I don't think this is going to happen - and if it is, certainly not
any time soon.
> 2) How do we monitor VM's that are in Cloudstack when they are in an isolated VLAN does anyone have a clever workaround?
I dont have a good answer for this at the moment
> 3) Has anyone developed a script for parsing and alerting on warning events in the management Log yet?
>
So you should check out Zenoss's CloudStack ZenPack [1] they poll the
listEvents API call and parse that for events.
Alternatively, you could configure log4j to send logs of warning or
higher to syslog - and thus on to some central logging facility like
logstash, greylog, or splunk.
[1] https://github.com/zenoss/ZenPacks.zenoss.CloudStack