Hi all,
I have a weird behaviour in Foreman, running a long blocking task will
influence the entire system and even simple/short task will fail.
For example:
I have a compute resource overseas and every time a open a VM located in
that resource, it takes minutes until the VM is completely loaded.
That is a problem but a bigger problem is while this task is running i have
no option to even click on "hosts" button because it will take forever
until something will move.
It seams like the blocking task has locked the DB and nothing else can be
done until it completed.
Guys, I need help to debug this problem and find ways to workaround or even
better, fix it.
···
On Monday, June 29, 2015 at 10:56:59 PM UTC+3, Avi Tal wrote:
>
> Hi all,
> I have a weird behaviour in Foreman, running a long blocking task will
> influence the entire system and even simple/short task will fail.
>
> For example:
> I have a compute resource overseas and every time a open a VM located in
> that resource, it takes minutes until the VM is completely loaded.
> That is a problem but a bigger problem is while this task is running i
> have no option to even click on "hosts" button because it will take forever
> until something will move.
>
> It seams like the blocking task has locked the DB and nothing else can be
> done until it completed.
>
> Guys, I need help to debug this problem and find ways to workaround or
> even better, fix it.
>
> Thanks
>
> BTW, Foreman version 1.8.2 on CentOS 6.5
>
>
>>
>> Hi all,
>> I have a weird behaviour in Foreman, running a long blocking task will
>> influence the entire system and even simple/short task will fail.
>>
>> For example:
>> I have a compute resource overseas and every time a open a VM located in
>> that resource, it takes minutes until the VM is completely loaded.
>> That is a problem but a bigger problem is while this task is running i
>> have no option to even click on "hosts" button because it will take forever
>> until something will move.
>>
>> It seams like the blocking task has locked the DB and nothing else can be
>> done until it completed.
>>
>> Guys, I need help to debug this problem and find ways to workaround or
>> even better, fix it.
>>
>
how do you run foreman? using passenger? how many workers?
Ohad
···
On Mon, Jun 29, 2015 at 11:14 PM, Avi Tal wrote:
> On Monday, June 29, 2015 at 10:56:59 PM UTC+3, Avi Tal wrote:
I am using the default configuration since I have never changed it.
So yes it is passenger, and I have no idea about the workers.
I'll appreciate your guidance for high scalable / performance multi
communication foreman.
Our work with foreman requires many communications in parallel with many
provision tasks.
On Monday, June 29, 2015 at 10:56:59 PM UTC+3, Avi Tal wrote:
Hi all,
I have a weird behaviour in Foreman, running a long blocking task will
influence the entire system and even simple/short task will fail.
For example:
I have a compute resource overseas and every time a open a VM located in
that resource, it takes minutes until the VM is completely loaded.
That is a problem but a bigger problem is while this task is running i
have no option to even click on “hosts” button because it will take forever
until something will move.
It seams like the blocking task has locked the DB and nothing else can
be done until it completed.
Guys, I need help to debug this problem and find ways to workaround or
even better, fix it.
how do you run foreman? using passenger? how many workers?
Ohad
Thanks
–
You received this message because you are subscribed to the Google Groups
"Foreman users" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to foreman-users+unsubscribe@googlegroups.com.
My foreman is VM centos 6.5 x86_64 with 4 CPUs and 8 G RAM
I have add the following lines to 05-foreman.conf (inside <VirtualHost
*:80>) PassengerMaxPoolSize 30 PassengerConcurrencyModel thread PassengerPreStart
How does it look?
···
On Tuesday, June 30, 2015 at 6:18:01 PM UTC+3, Avi Tal wrote:
>
> I am using the default configuration since I have never changed it.
> So yes it is passenger, and I have no idea about the workers.
> I'll appreciate your guidance for high scalable / performance multi
> communication foreman.
> Our work with foreman requires many communications in parallel with many
> provision tasks.
>
> On Tue, Jun 30, 2015, 09:51 Ohad Levy wrote:
>
>> On Mon, Jun 29, 2015 at 11:14 PM, Avi Tal wrote:
>>
>>> BTW, Foreman version 1.8.2 on CentOS 6.5
>>>
>>>
>>> On Monday, June 29, 2015 at 10:56:59 PM UTC+3, Avi Tal wrote:
>>>>
>>>> Hi all,
>>>> I have a weird behaviour in Foreman, running a long blocking task will
>>>> influence the entire system and even simple/short task will fail.
>>>>
>>>> For example:
>>>> I have a compute resource overseas and every time a open a VM located
>>>> in that resource, it takes minutes until the VM is completely loaded.
>>>> That is a problem but a bigger problem is while this task is running i
>>>> have no option to even click on "hosts" button because it will take forever
>>>> until something will move.
>>>>
>>>> It seams like the blocking task has locked the DB and nothing else can
>>>> be done until it completed.
>>>>
>>>> Guys, I need help to debug this problem and find ways to workaround or
>>>> even better, fix it.
>>>>
>>>
>> how do you run foreman? using passenger? how many workers?
>>
>> Ohad
>>
>>>
>>>> Thanks
>>>>
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "Foreman users" group.
>>> To unsubscribe from this group and stop receiving emails from it, send
>>> an email to foreman-users+unsubscribe@googlegroups.com.
>>>
>>
>>> To post to this group, send email to foreman-users@googlegroups.com.
>>> Visit this group at http://groups.google.com/group/foreman-users.
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>> --
>> You received this message because you are subscribed to a topic in the
>> Google Groups "Foreman users" group.
>> To unsubscribe from this topic, visit
>> https://groups.google.com/d/topic/foreman-users/GOBc6dpx7KY/unsubscribe.
>> To unsubscribe from this group and all its topics, send an email to
>> foreman-users+unsubscribe@googlegroups.com.
>> To post to this group, send email to foreman-users@googlegroups.com.
>> Visit this group at http://groups.google.com/group/foreman-users.
>> For more options, visit https://groups.google.com/d/optout.
>>
>
Actually *PassengerConcurrencyModel *and *PassengerPreStart *caused an
error so i have only added PassengerMaxPoolSize 30
···
On Wednesday, July 1, 2015 at 11:46:05 AM UTC+3, Avi Tal wrote:
>
> My foreman is VM *centos 6.5 x86_64 with 4 CPUs and 8 G RAM*
>
> I have add the following lines to 05-foreman.conf (inside *:80>)
> *PassengerMaxPoolSize 30*
> *PassengerConcurrencyModel thread*
> *PassengerPreStart*
>
> How does it look?
>
>
> On Tuesday, June 30, 2015 at 6:18:01 PM UTC+3, Avi Tal wrote:
>>
>> I am using the default configuration since I have never changed it.
>> So yes it is passenger, and I have no idea about the workers.
>> I'll appreciate your guidance for high scalable / performance multi
>> communication foreman.
>> Our work with foreman requires many communications in parallel with many
>> provision tasks.
>>
>> On Tue, Jun 30, 2015, 09:51 Ohad Levy wrote:
>>
>>> On Mon, Jun 29, 2015 at 11:14 PM, Avi Tal wrote:
>>>
>>>> BTW, Foreman version 1.8.2 on CentOS 6.5
>>>>
>>>>
>>>> On Monday, June 29, 2015 at 10:56:59 PM UTC+3, Avi Tal wrote:
>>>>>
>>>>> Hi all,
>>>>> I have a weird behaviour in Foreman, running a long blocking task will
>>>>> influence the entire system and even simple/short task will fail.
>>>>>
>>>>> For example:
>>>>> I have a compute resource overseas and every time a open a VM located
>>>>> in that resource, it takes minutes until the VM is completely loaded.
>>>>> That is a problem but a bigger problem is while this task is running i
>>>>> have no option to even click on "hosts" button because it will take forever
>>>>> until something will move.
>>>>>
>>>>> It seams like the blocking task has locked the DB and nothing else can
>>>>> be done until it completed.
>>>>>
>>>>> Guys, I need help to debug this problem and find ways to workaround or
>>>>> even better, fix it.
>>>>>
>>>>
>>> how do you run foreman? using passenger? how many workers?
>>>
>>> Ohad
>>>
>>>>
>>>>> Thanks
>>>>>
>>>> --
>>>> You received this message because you are subscribed to the Google
>>>> Groups "Foreman users" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>> an email to foreman-users+unsubscribe@googlegroups.com.
>>>>
>>>
>>>> To post to this group, send email to foreman-users@googlegroups.com.
>>>> Visit this group at http://groups.google.com/group/foreman-users.
>>>> For more options, visit https://groups.google.com/d/optout.
>>>>
>>> --
>>> You received this message because you are subscribed to a topic in the
>>> Google Groups "Foreman users" group.
>>> To unsubscribe from this topic, visit
>>> https://groups.google.com/d/topic/foreman-users/GOBc6dpx7KY/unsubscribe.
>>> To unsubscribe from this group and all its topics, send an email to
>>> foreman-users+unsubscribe@googlegroups.com.
>>> To post to this group, send email to foreman-users@googlegroups.com.
>>> Visit this group at http://groups.google.com/group/foreman-users.
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>
how many hosts you are scaling to? i am researching foreman scalability for
a largescale number of nodes.
···
On 1 July 2015 at 11:54, Avi Tal wrote:
Actually *PassengerConcurrencyModel *and *PassengerPreStart *caused an
error so i have only added PassengerMaxPoolSize 30
On Wednesday, July 1, 2015 at 11:46:05 AM UTC+3, Avi Tal wrote:
My foreman is VM centos 6.5 x86_64 with 4 CPUs and 8 G RAM
I have add the following lines to 05-foreman.conf (inside <VirtualHost
*:80>) PassengerMaxPoolSize 30 PassengerConcurrencyModel thread PassengerPreStart
How does it look?
On Tuesday, June 30, 2015 at 6:18:01 PM UTC+3, Avi Tal wrote:
I am using the default configuration since I have never changed it.
So yes it is passenger, and I have no idea about the workers.
I’ll appreciate your guidance for high scalable / performance multi
communication foreman.
Our work with foreman requires many communications in parallel with many
provision tasks.
On Monday, June 29, 2015 at 10:56:59 PM UTC+3, Avi Tal wrote:
Hi all,
I have a weird behaviour in Foreman, running a long blocking task
will influence the entire system and even simple/short task will fail.
For example:
I have a compute resource overseas and every time a open a VM located
in that resource, it takes minutes until the VM is completely loaded.
That is a problem but a bigger problem is while this task is running
i have no option to even click on “hosts” button because it will take
forever until something will move.
It seams like the blocking task has locked the DB and nothing else
can be done until it completed.
Guys, I need help to debug this problem and find ways to workaround
or even better, fix it.
how do you run foreman? using passenger? how many workers?
Ohad
Thanks
–
You received this message because you are subscribed to the Google
Groups “Foreman users” group.
To unsubscribe from this group and stop receiving emails from it, send
an email to foreman-users+unsubscribe@googlegroups.com.
Hi,
Well our firmware creates hosts intensively, growing to around 300 virtual
hosts and 30 physical.
The problem still exists and feels like if the connection to vcenter gets a
bit slow influence other connections even though not all related to
vcenter.
Our framework currently suffer from the latency of foreman and many tasks
just fail due to timeout.
I recommend:
Raising the multi connection mechanism in foreman api.
Implement same structure on smart proxy, instead of having only one
connection.
Fix the coupling of tasks within foreman compute resource by incapsulate
tasks related resources.
Maybe even redesign the reactor/proactor threading mechanism in foreman
rest api.
I'll be happy to assist if needed.
Thanks
···
On Sat, Aug 22, 2015, 20:04 Walid wrote:
Hi Avi,
how many hosts you are scaling to? i am researching foreman scalability
for a largescale number of nodes.
Actually *PassengerConcurrencyModel *and *PassengerPreStart *caused an
error so i have only added PassengerMaxPoolSize 30
On Wednesday, July 1, 2015 at 11:46:05 AM UTC+3, Avi Tal wrote:
My foreman is VM centos 6.5 x86_64 with 4 CPUs and 8 G RAM
I have add the following lines to 05-foreman.conf (inside <VirtualHost
*:80>) PassengerMaxPoolSize 30 PassengerConcurrencyModel thread PassengerPreStart
How does it look?
On Tuesday, June 30, 2015 at 6:18:01 PM UTC+3, Avi Tal wrote:
I am using the default configuration since I have never changed it.
So yes it is passenger, and I have no idea about the workers.
I’ll appreciate your guidance for high scalable / performance multi
communication foreman.
Our work with foreman requires many communications in parallel with
many provision tasks.
On Monday, June 29, 2015 at 10:56:59 PM UTC+3, Avi Tal wrote:
Hi all,
I have a weird behaviour in Foreman, running a long blocking task
will influence the entire system and even simple/short task will fail.
For example:
I have a compute resource overseas and every time a open a VM
located in that resource, it takes minutes until the VM is completely
loaded.
That is a problem but a bigger problem is while this task is running
i have no option to even click on “hosts” button because it will take
forever until something will move.
It seams like the blocking task has locked the DB and nothing else
can be done until it completed.
Guys, I need help to debug this problem and find ways to workaround
or even better, fix it.
how do you run foreman? using passenger? how many workers?
Ohad
Thanks
–
You received this message because you are subscribed to the Google
Groups “Foreman users” group.
To unsubscribe from this group and stop receiving emails from it,
send an email to foreman-users+unsubscribe@googlegroups.com.
thanks Avi, my concern is more with after provisioning, would foreman scale
or not to support 1000s of nodes, reporting on puppet agent runs, and how
many nodes would one smart proxy handle, and how to scale the smart proxies?
thank you for your reply.
···
On 22 August 2015 at 21:13, Avi Tal wrote:
Hi,
Well our firmware creates hosts intensively, growing to around 300 virtual
hosts and 30 physical.
The problem still exists and feels like if the connection to vcenter gets
a bit slow influence other connections even though not all related to
vcenter.
Our framework currently suffer from the latency of foreman and many tasks
just fail due to timeout.
I recommend:
Raising the multi connection mechanism in foreman api.
Implement same structure on smart proxy, instead of having only one
connection.
Fix the coupling of tasks within foreman compute resource by
incapsulate tasks related resources.
Maybe even redesign the reactor/proactor threading mechanism in foreman
rest api.
Actually *PassengerConcurrencyModel *and *PassengerPreStart *caused an
error so i have only added PassengerMaxPoolSize 30
On Wednesday, July 1, 2015 at 11:46:05 AM UTC+3, Avi Tal wrote:
My foreman is VM centos 6.5 x86_64 with 4 CPUs and 8 G RAM
I have add the following lines to 05-foreman.conf (inside <VirtualHost
*:80>) PassengerMaxPoolSize 30 PassengerConcurrencyModel thread PassengerPreStart
How does it look?
On Tuesday, June 30, 2015 at 6:18:01 PM UTC+3, Avi Tal wrote:
I am using the default configuration since I have never changed it.
So yes it is passenger, and I have no idea about the workers.
I’ll appreciate your guidance for high scalable / performance multi
communication foreman.
Our work with foreman requires many communications in parallel with
many provision tasks.
On Monday, June 29, 2015 at 10:56:59 PM UTC+3, Avi Tal wrote:
Hi all,
I have a weird behaviour in Foreman, running a long blocking task
will influence the entire system and even simple/short task will fail.
For example:
I have a compute resource overseas and every time a open a VM
located in that resource, it takes minutes until the VM is completely
loaded.
That is a problem but a bigger problem is while this task is
running i have no option to even click on “hosts” button because it will
take forever until something will move.
It seams like the blocking task has locked the DB and nothing else
can be done until it completed.
Guys, I need help to debug this problem and find ways to workaround
or even better, fix it.
how do you run foreman? using passenger? how many workers?
Ohad
Thanks
–
You received this message because you are subscribed to the Google
Groups “Foreman users” group.
To unsubscribe from this group and stop receiving emails from it,
send an email to foreman-users+unsubscribe@googlegroups.com.
> thanks Avi, my concern is more with after provisioning, would foreman
> scale or not to support 1000s of nodes, reporting on puppet agent runs, and
> how many nodes would one smart proxy handle, and how to scale the smart
> proxies?
>
Our largest installation (that we know of) is around 100K hosts, and many
others in the area of 50K hosts as well.
Hi,
Well our firmware creates hosts intensively, growing to around 300
virtual hosts and 30 physical.
The problem still exists and feels like if the connection to vcenter gets
a bit slow influence other connections even though not all related to
vcenter.
Our framework currently suffer from the latency of foreman and many tasks
just fail due to timeout.
I recommend:
Raising the multi connection mechanism in foreman api.
Implement same structure on smart proxy, instead of having only one
connection.
Fix the coupling of tasks within foreman compute resource by
incapsulate tasks related resources.
Maybe even redesign the reactor/proactor threading mechanism in
foreman rest api.
Actually *PassengerConcurrencyModel *and *PassengerPreStart *caused an
error so i have only added PassengerMaxPoolSize 30
On Wednesday, July 1, 2015 at 11:46:05 AM UTC+3, Avi Tal wrote:
My foreman is VM centos 6.5 x86_64 with 4 CPUs and 8 G RAM
I have add the following lines to 05-foreman.conf (inside <VirtualHost
*:80>) PassengerMaxPoolSize 30 PassengerConcurrencyModel thread PassengerPreStart
How does it look?
On Tuesday, June 30, 2015 at 6:18:01 PM UTC+3, Avi Tal wrote:
I am using the default configuration since I have never changed it.
So yes it is passenger, and I have no idea about the workers.
I’ll appreciate your guidance for high scalable / performance multi
communication foreman.
Our work with foreman requires many communications in parallel with
many provision tasks.
On Monday, June 29, 2015 at 10:56:59 PM UTC+3, Avi Tal wrote:
Hi all,
I have a weird behaviour in Foreman, running a long blocking task
will influence the entire system and even simple/short task will fail.
For example:
I have a compute resource overseas and every time a open a VM
located in that resource, it takes minutes until the VM is completely
loaded.
That is a problem but a bigger problem is while this task is
running i have no option to even click on “hosts” button because it will
take forever until something will move.
It seams like the blocking task has locked the DB and nothing else
can be done until it completed.
Guys, I need help to debug this problem and find ways to
workaround or even better, fix it.
how do you run foreman? using passenger? how many workers?
Ohad
Thanks
–
You received this message because you are subscribed to the Google
Groups “Foreman users” group.
To unsubscribe from this group and stop receiving emails from it,
send an email to foreman-users+unsubscribe@googlegroups.com.
so how do they manage to scale the puppet configuration and reporting, just
by standard smartproxy? or they have to load balance puppet masters in each
smartproxy some how, any urls, or hints are definitely welcome
thanks Avi, my concern is more with after provisioning, would foreman
scale or not to support 1000s of nodes, reporting on puppet agent runs, and
how many nodes would one smart proxy handle, and how to scale the smart
proxies?
Our largest installation (that we know of) is around 100K hosts, and many
others in the area of 50K hosts as well.
Hi,
Well our firmware creates hosts intensively, growing to around 300
virtual hosts and 30 physical.
The problem still exists and feels like if the connection to vcenter
gets a bit slow influence other connections even though not all related to
vcenter.
Our framework currently suffer from the latency of foreman and many
tasks just fail due to timeout.
I recommend:
Raising the multi connection mechanism in foreman api.
Implement same structure on smart proxy, instead of having only one
connection.
Fix the coupling of tasks within foreman compute resource by
incapsulate tasks related resources.
Maybe even redesign the reactor/proactor threading mechanism in
foreman rest api.
Actually *PassengerConcurrencyModel *and *PassengerPreStart *caused
an error so i have only added PassengerMaxPoolSize 30
On Wednesday, July 1, 2015 at 11:46:05 AM UTC+3, Avi Tal wrote:
My foreman is VM centos 6.5 x86_64 with 4 CPUs and 8 G RAM
I have add the following lines to 05-foreman.conf
(inside <VirtualHost *:80>) PassengerMaxPoolSize 30 PassengerConcurrencyModel thread PassengerPreStart
How does it look?
On Tuesday, June 30, 2015 at 6:18:01 PM UTC+3, Avi Tal wrote:
I am using the default configuration since I have never changed it.
So yes it is passenger, and I have no idea about the workers.
I’ll appreciate your guidance for high scalable / performance multi
communication foreman.
Our work with foreman requires many communications in parallel with
many provision tasks.
On Mon, Jun 29, 2015 at 11:14 PM, Avi Tal avi3tal@gmail.com >>>>>>>> wrote:
BTW, Foreman version 1.8.2 on CentOS 6.5
On Monday, June 29, 2015 at 10:56:59 PM UTC+3, Avi Tal wrote:
Hi all,
I have a weird behaviour in Foreman, running a long blocking task
will influence the entire system and even simple/short task will fail.
For example:
I have a compute resource overseas and every time a open a VM
located in that resource, it takes minutes until the VM is completely
loaded.
That is a problem but a bigger problem is while this task is
running i have no option to even click on “hosts” button because it will
take forever until something will move.
It seams like the blocking task has locked the DB and nothing
else can be done until it completed.
Guys, I need help to debug this problem and find ways to
workaround or even better, fix it.
how do you run foreman? using passenger? how many workers?
Ohad
Thanks
–
You received this message because you are subscribed to the Google
Groups “Foreman users” group.
To unsubscribe from this group and stop receiving emails from it,
send an email to foreman-users+unsubscribe@googlegroups.com.
> Hi Ohad,
>
> so how do they manage to scale the puppet configuration and reporting,
> just by standard smartproxy? or they have to load balance puppet masters in
> each smartproxy some how, any urls, or hints are definitely welcome
>
while this scale is probably different from what most people need, have a
look at [1]
before watching, I would only say that simple load balancer + multiple
foreman instances (with a single/replicated / single master db) goes a long
way to achieve thousands of clients.
thanks Avi, my concern is more with after provisioning, would foreman
scale or not to support 1000s of nodes, reporting on puppet agent runs, and
how many nodes would one smart proxy handle, and how to scale the smart
proxies?
Our largest installation (that we know of) is around 100K hosts, and many
others in the area of 50K hosts as well.
Hi,
Well our firmware creates hosts intensively, growing to around 300
virtual hosts and 30 physical.
The problem still exists and feels like if the connection to vcenter
gets a bit slow influence other connections even though not all related to
vcenter.
Our framework currently suffer from the latency of foreman and many
tasks just fail due to timeout.
I recommend:
Raising the multi connection mechanism in foreman api.
Implement same structure on smart proxy, instead of having only one
connection.
Fix the coupling of tasks within foreman compute resource by
incapsulate tasks related resources.
Maybe even redesign the reactor/proactor threading mechanism in
foreman rest api.
Actually *PassengerConcurrencyModel *and *PassengerPreStart *caused
an error so i have only added PassengerMaxPoolSize 30
On Wednesday, July 1, 2015 at 11:46:05 AM UTC+3, Avi Tal wrote:
My foreman is VM centos 6.5 x86_64 with 4 CPUs and 8 G RAM
I have add the following lines to 05-foreman.conf
(inside <VirtualHost *:80>) PassengerMaxPoolSize 30 PassengerConcurrencyModel thread PassengerPreStart
How does it look?
On Tuesday, June 30, 2015 at 6:18:01 PM UTC+3, Avi Tal wrote:
I am using the default configuration since I have never changed it.
So yes it is passenger, and I have no idea about the workers.
I’ll appreciate your guidance for high scalable / performance multi
communication foreman.
Our work with foreman requires many communications in parallel with
many provision tasks.
On Mon, Jun 29, 2015 at 11:14 PM, Avi Tal avi3tal@gmail.com >>>>>>>>> wrote:
BTW, Foreman version 1.8.2 on CentOS 6.5
On Monday, June 29, 2015 at 10:56:59 PM UTC+3, Avi Tal wrote:
Hi all,
I have a weird behaviour in Foreman, running a long blocking
task will influence the entire system and even simple/short task will fail.
For example:
I have a compute resource overseas and every time a open a VM
located in that resource, it takes minutes until the VM is completely
loaded.
That is a problem but a bigger problem is while this task is
running i have no option to even click on “hosts” button because it will
take forever until something will move.
It seams like the blocking task has locked the DB and nothing
else can be done until it completed.
Guys, I need help to debug this problem and find ways to
workaround or even better, fix it.
how do you run foreman? using passenger? how many workers?
Ohad
Thanks
–
You received this message because you are subscribed to the
Google Groups “Foreman users” group.
To unsubscribe from this group and stop receiving emails from it,
send an email to foreman-users+unsubscribe@googlegroups.com.
so how do they manage to scale the puppet configuration and reporting,
just by standard smartproxy? or they have to load balance puppet masters in
each smartproxy some how, any urls, or hints are definitely welcome
while this scale is probably different from what most people need, have a
look at [1]
before watching, I would only say that simple load balancer + multiple
foreman instances (with a single/replicated / single master db) goes a long
way to achieve thousands of clients.
thanks Avi, my concern is more with after provisioning, would foreman
scale or not to support 1000s of nodes, reporting on puppet agent runs, and
how many nodes would one smart proxy handle, and how to scale the smart
proxies?
Our largest installation (that we know of) is around 100K hosts, and
many others in the area of 50K hosts as well.
Hi,
Well our firmware creates hosts intensively, growing to around 300
virtual hosts and 30 physical.
The problem still exists and feels like if the connection to vcenter
gets a bit slow influence other connections even though not all related to
vcenter.
Our framework currently suffer from the latency of foreman and many
tasks just fail due to timeout.
I recommend:
Raising the multi connection mechanism in foreman api.
Implement same structure on smart proxy, instead of having only one
connection.
Fix the coupling of tasks within foreman compute resource by
incapsulate tasks related resources.
Maybe even redesign the reactor/proactor threading mechanism in
foreman rest api.
Actually *PassengerConcurrencyModel *and *PassengerPreStart *caused
an error so i have only added PassengerMaxPoolSize 30
On Wednesday, July 1, 2015 at 11:46:05 AM UTC+3, Avi Tal wrote:
My foreman is VM centos 6.5 x86_64 with 4 CPUs and 8 G RAM
I have add the following lines to 05-foreman.conf
(inside <VirtualHost *:80>) PassengerMaxPoolSize 30 PassengerConcurrencyModel thread PassengerPreStart
How does it look?
On Tuesday, June 30, 2015 at 6:18:01 PM UTC+3, Avi Tal wrote:
I am using the default configuration since I have never changed
it.
So yes it is passenger, and I have no idea about the workers.
I’ll appreciate your guidance for high scalable / performance
multi communication foreman.
Our work with foreman requires many communications in parallel
with many provision tasks.
On Mon, Jun 29, 2015 at 11:14 PM, Avi Tal avi3tal@gmail.com >>>>>>>>>> wrote:
BTW, Foreman version 1.8.2 on CentOS 6.5
On Monday, June 29, 2015 at 10:56:59 PM UTC+3, Avi Tal wrote:
Hi all,
I have a weird behaviour in Foreman, running a long blocking
task will influence the entire system and even simple/short task will fail.
For example:
I have a compute resource overseas and every time a open a VM
located in that resource, it takes minutes until the VM is completely
loaded.
That is a problem but a bigger problem is while this task is
running i have no option to even click on “hosts” button because it will
take forever until something will move.
It seams like the blocking task has locked the DB and nothing
else can be done until it completed.
Guys, I need help to debug this problem and find ways to
workaround or even better, fix it.
how do you run foreman? using passenger? how many workers?
Ohad
Thanks
–
You received this message because you are subscribed to the
Google Groups “Foreman users” group.
To unsubscribe from this group and stop receiving emails from
it, send an email to foreman-users+unsubscribe@googlegroups.com.
