You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@cloudstack.apache.org by Piotr Pisz <pi...@piszki.pl> on 2019/09/17 05:59:33 UTC
disk total vs disk allocated
Hi all,
I have a strange situation, we have a CephFS share mounted as SharedMountPoint.
CS shows Disk Total as 12T (that's ok), while Disk Allocated shows like 11.8T (it's not ok, disk is 50% full).
How can We diagnose the cause?
Regards,
Piotr
Re: disk total vs disk allocated
Posted by Thomas Joseph <th...@gmail.com>.
Hello Piotr,
Are you using thin provisioning for disk offerings?
With regards
Thomas
On Tue, 17 Sep 2019, 7:00 am Piotr Pisz, <pi...@piszki.pl> wrote:
> Hi all,
>
> I have a strange situation, we have a CephFS share mounted as
> SharedMountPoint.
> CS shows Disk Total as 12T (that's ok), while Disk Allocated shows like
> 11.8T (it's not ok, disk is 50% full).
> How can We diagnose the cause?
>
> Regards,
> Piotr
>
>
Re: disk total vs disk allocated
Posted by Andrija Panic <an...@gmail.com>.
Wondering if updating DB directly would have any effect on the calculations
- It will probably happen that this pool "detail" (overprovisioning factor)
is not taken into consideration at all, even if the DB record is there.
Try adding a new row to the storage_pool_details table - i.e.
INSERT INTO `cloud`.`storage_pool_details` (`id`, `pool_id`, `name`,
`value`, `display`) VALUES ('9', '2', 'storage.overprovisioning.factor',
'2', '1');
Obviously, change your pool_id and the id as needed in that table.
Andrija
On Tue, Sep 17, 2019, 10:46 Piotr Pisz <pi...@piszki.pl> wrote:
> We've set the overprovisioning to 2, and clearly doesn't work on
> SharedMountPoint :(
>
>
>
> “Unable to update storage pool with id 19. Overprovision not supported
> for SharedMountPoint”
>
>
>
> Piotr
>
>
>
>
>
> *From:* Andrija Panic <an...@gmail.com>
> *Sent:* Tuesday, September 17, 2019 9:15 AM
> *To:* users <us...@cloudstack.apache.org>; Piotr Pisz <pi...@piszki.pl>
> *Subject:* Re: disk total vs disk allocated
>
>
>
> So, that sounds OK I guess.
>
>
>
> Can you try setting the overprovisioning on that Primary Storage
> specifically? I can't really say if storage overprovisioning is supported
> for shared mount point (it is supported for NFS, local and probably RBD) -
> though that is only the code logic.
>
>
>
> Andrija
>
>
>
> On Tue, Sep 17, 2019, 08:50 Piotr Pisz <pi...@piszki.pl> wrote:
>
> Thank you everyone, I didn't really realize that my friends made so many
> thin disks.
> The storage.alloction.overprovisioning.factor parameter could be useful,
> what is the point of using thin disks if the capacity of the pool cannot be
> exceeded?
>
> MariaDB [cloud]> select round(size/1024/1024/1024, 2) as "size [GB]",
> state from volumes where pool_id = 19 and removed is NULL;
> +-----------+-------+
> | size [GB] | state |
> +-----------+-------+
> | 2048.00 | Ready |
> | 0.36 | Ready |
> | 1024.00 | Ready |
> | 1024.00 | Ready |
> | 1024.00 | Ready |
> | 1024.00 | Ready |
> | 1024.00 | Ready |
> | 1024.00 | Ready |
> | 1024.00 | Ready |
> | 30.00 | Ready |
> | 100.00 | Ready |
> | 50.00 | Ready |
> | 60.00 | Ready |
> | 50.00 | Ready |
> | 50.00 | Ready |
> | 50.00 | Ready |
> | 100.00 | Ready |
> | 50.00 | Ready |
> | 100.00 | Ready |
> | 500.00 | Ready |
> | 250.00 | Ready |
> | 250.00 | Ready |
> | 250.00 | Ready |
> | 250.00 | Ready |
> | 250.00 | Ready |
> | 40.00 | Ready |
> | 100.00 | Ready |
> +-----------+-------+
> 27 rows in set (0.001 sec)
>
> Regards,
> Piotr
>
>
> -----Original Message-----
> From: Andrija Panic <an...@gmail.com>
> Sent: Tuesday, September 17, 2019 8:10 AM
> To: users <us...@cloudstack.apache.org>; Piotr Pisz <pi...@piszki.pl>
> Subject: Re: disk total vs disk allocated
>
> Wow.. . That's the community spirit - 3 replies in 8min from the original
> email!
>
> On Tue, Sep 17, 2019, 08:07 Andrija Panic <an...@gmail.com> wrote:
>
> > Hi Piotr,
> >
> > Allocated should be the sum of all disks virtual size from the DB,
> > "volumes" table.
> >
> > Can you do select sum "size" from that table, where pool_id=xxx and
> > removed is NULL (on my mobile, can't remember the name of the column
> > in the table, probability just "size").
> >
> > More importantly, I'm wondering how is this working for you? Did you
> > have time to test that setup "properly"? Any feedbacks?
> >
> > Best,
> > Andrija
> >
> > On Tue, Sep 17, 2019, 08:00 Piotr Pisz <pi...@piszki.pl> wrote:
> >
> >> Hi all,
> >>
> >> I have a strange situation, we have a CephFS share mounted as
> >> SharedMountPoint.
> >> CS shows Disk Total as 12T (that's ok), while Disk Allocated shows
> >> like 11.8T (it's not ok, disk is 50% full).
> >> How can We diagnose the cause?
> >>
> >> Regards,
> >> Piotr
> >>
> >>
>
>
RE: disk total vs disk allocated
Posted by Piotr Pisz <pi...@piszki.pl>.
We've set the overprovisioning to 2, and clearly doesn't work on SharedMountPoint :(
“Unable to update storage pool with id 19. Overprovision not supported for SharedMountPoint”
Piotr
From: Andrija Panic <an...@gmail.com>
Sent: Tuesday, September 17, 2019 9:15 AM
To: users <us...@cloudstack.apache.org>; Piotr Pisz <pi...@piszki.pl>
Subject: Re: disk total vs disk allocated
So, that sounds OK I guess.
Can you try setting the overprovisioning on that Primary Storage specifically? I can't really say if storage overprovisioning is supported for shared mount point (it is supported for NFS, local and probably RBD) - though that is only the code logic.
Andrija
On Tue, Sep 17, 2019, 08:50 Piotr Pisz <piotr@piszki.pl <ma...@piszki.pl> > wrote:
Thank you everyone, I didn't really realize that my friends made so many thin disks.
The storage.alloction.overprovisioning.factor parameter could be useful, what is the point of using thin disks if the capacity of the pool cannot be exceeded?
MariaDB [cloud]> select round(size/1024/1024/1024, 2) as "size [GB]", state from volumes where pool_id = 19 and removed is NULL;
+-----------+-------+
| size [GB] | state |
+-----------+-------+
| 2048.00 | Ready |
| 0.36 | Ready |
| 1024.00 | Ready |
| 1024.00 | Ready |
| 1024.00 | Ready |
| 1024.00 | Ready |
| 1024.00 | Ready |
| 1024.00 | Ready |
| 1024.00 | Ready |
| 30.00 | Ready |
| 100.00 | Ready |
| 50.00 | Ready |
| 60.00 | Ready |
| 50.00 | Ready |
| 50.00 | Ready |
| 50.00 | Ready |
| 100.00 | Ready |
| 50.00 | Ready |
| 100.00 | Ready |
| 500.00 | Ready |
| 250.00 | Ready |
| 250.00 | Ready |
| 250.00 | Ready |
| 250.00 | Ready |
| 250.00 | Ready |
| 40.00 | Ready |
| 100.00 | Ready |
+-----------+-------+
27 rows in set (0.001 sec)
Regards,
Piotr
-----Original Message-----
From: Andrija Panic <andrija.panic@gmail.com <ma...@gmail.com> >
Sent: Tuesday, September 17, 2019 8:10 AM
To: users <users@cloudstack.apache.org <ma...@cloudstack.apache.org> >; Piotr Pisz <piotr@piszki.pl <ma...@piszki.pl> >
Subject: Re: disk total vs disk allocated
Wow.. . That's the community spirit - 3 replies in 8min from the original email!
On Tue, Sep 17, 2019, 08:07 Andrija Panic <andrija.panic@gmail.com <ma...@gmail.com> > wrote:
> Hi Piotr,
>
> Allocated should be the sum of all disks virtual size from the DB,
> "volumes" table.
>
> Can you do select sum "size" from that table, where pool_id=xxx and
> removed is NULL (on my mobile, can't remember the name of the column
> in the table, probability just "size").
>
> More importantly, I'm wondering how is this working for you? Did you
> have time to test that setup "properly"? Any feedbacks?
>
> Best,
> Andrija
>
> On Tue, Sep 17, 2019, 08:00 Piotr Pisz <piotr@piszki.pl <ma...@piszki.pl> > wrote:
>
>> Hi all,
>>
>> I have a strange situation, we have a CephFS share mounted as
>> SharedMountPoint.
>> CS shows Disk Total as 12T (that's ok), while Disk Allocated shows
>> like 11.8T (it's not ok, disk is 50% full).
>> How can We diagnose the cause?
>>
>> Regards,
>> Piotr
>>
>>
Re: disk total vs disk allocated
Posted by Andrija Panic <an...@gmail.com>.
So, that sounds OK I guess.
Can you try setting the overprovisioning on that Primary Storage
specifically? I can't really say if storage overprovisioning is supported
for shared mount point (it is supported for NFS, local and probably RBD) -
though that is only the code logic.
Andrija
On Tue, Sep 17, 2019, 08:50 Piotr Pisz <pi...@piszki.pl> wrote:
> Thank you everyone, I didn't really realize that my friends made so many
> thin disks.
> The storage.alloction.overprovisioning.factor parameter could be useful,
> what is the point of using thin disks if the capacity of the pool cannot be
> exceeded?
>
> MariaDB [cloud]> select round(size/1024/1024/1024, 2) as "size [GB]",
> state from volumes where pool_id = 19 and removed is NULL;
> +-----------+-------+
> | size [GB] | state |
> +-----------+-------+
> | 2048.00 | Ready |
> | 0.36 | Ready |
> | 1024.00 | Ready |
> | 1024.00 | Ready |
> | 1024.00 | Ready |
> | 1024.00 | Ready |
> | 1024.00 | Ready |
> | 1024.00 | Ready |
> | 1024.00 | Ready |
> | 30.00 | Ready |
> | 100.00 | Ready |
> | 50.00 | Ready |
> | 60.00 | Ready |
> | 50.00 | Ready |
> | 50.00 | Ready |
> | 50.00 | Ready |
> | 100.00 | Ready |
> | 50.00 | Ready |
> | 100.00 | Ready |
> | 500.00 | Ready |
> | 250.00 | Ready |
> | 250.00 | Ready |
> | 250.00 | Ready |
> | 250.00 | Ready |
> | 250.00 | Ready |
> | 40.00 | Ready |
> | 100.00 | Ready |
> +-----------+-------+
> 27 rows in set (0.001 sec)
>
> Regards,
> Piotr
>
>
> -----Original Message-----
> From: Andrija Panic <an...@gmail.com>
> Sent: Tuesday, September 17, 2019 8:10 AM
> To: users <us...@cloudstack.apache.org>; Piotr Pisz <pi...@piszki.pl>
> Subject: Re: disk total vs disk allocated
>
> Wow.. . That's the community spirit - 3 replies in 8min from the original
> email!
>
> On Tue, Sep 17, 2019, 08:07 Andrija Panic <an...@gmail.com> wrote:
>
> > Hi Piotr,
> >
> > Allocated should be the sum of all disks virtual size from the DB,
> > "volumes" table.
> >
> > Can you do select sum "size" from that table, where pool_id=xxx and
> > removed is NULL (on my mobile, can't remember the name of the column
> > in the table, probability just "size").
> >
> > More importantly, I'm wondering how is this working for you? Did you
> > have time to test that setup "properly"? Any feedbacks?
> >
> > Best,
> > Andrija
> >
> > On Tue, Sep 17, 2019, 08:00 Piotr Pisz <pi...@piszki.pl> wrote:
> >
> >> Hi all,
> >>
> >> I have a strange situation, we have a CephFS share mounted as
> >> SharedMountPoint.
> >> CS shows Disk Total as 12T (that's ok), while Disk Allocated shows
> >> like 11.8T (it's not ok, disk is 50% full).
> >> How can We diagnose the cause?
> >>
> >> Regards,
> >> Piotr
> >>
> >>
>
>
RE: disk total vs disk allocated
Posted by Piotr Pisz <pi...@piszki.pl>.
Thank you everyone, I didn't really realize that my friends made so many thin disks.
The storage.alloction.overprovisioning.factor parameter could be useful, what is the point of using thin disks if the capacity of the pool cannot be exceeded?
MariaDB [cloud]> select round(size/1024/1024/1024, 2) as "size [GB]", state from volumes where pool_id = 19 and removed is NULL;
+-----------+-------+
| size [GB] | state |
+-----------+-------+
| 2048.00 | Ready |
| 0.36 | Ready |
| 1024.00 | Ready |
| 1024.00 | Ready |
| 1024.00 | Ready |
| 1024.00 | Ready |
| 1024.00 | Ready |
| 1024.00 | Ready |
| 1024.00 | Ready |
| 30.00 | Ready |
| 100.00 | Ready |
| 50.00 | Ready |
| 60.00 | Ready |
| 50.00 | Ready |
| 50.00 | Ready |
| 50.00 | Ready |
| 100.00 | Ready |
| 50.00 | Ready |
| 100.00 | Ready |
| 500.00 | Ready |
| 250.00 | Ready |
| 250.00 | Ready |
| 250.00 | Ready |
| 250.00 | Ready |
| 250.00 | Ready |
| 40.00 | Ready |
| 100.00 | Ready |
+-----------+-------+
27 rows in set (0.001 sec)
Regards,
Piotr
-----Original Message-----
From: Andrija Panic <an...@gmail.com>
Sent: Tuesday, September 17, 2019 8:10 AM
To: users <us...@cloudstack.apache.org>; Piotr Pisz <pi...@piszki.pl>
Subject: Re: disk total vs disk allocated
Wow.. . That's the community spirit - 3 replies in 8min from the original email!
On Tue, Sep 17, 2019, 08:07 Andrija Panic <an...@gmail.com> wrote:
> Hi Piotr,
>
> Allocated should be the sum of all disks virtual size from the DB,
> "volumes" table.
>
> Can you do select sum "size" from that table, where pool_id=xxx and
> removed is NULL (on my mobile, can't remember the name of the column
> in the table, probability just "size").
>
> More importantly, I'm wondering how is this working for you? Did you
> have time to test that setup "properly"? Any feedbacks?
>
> Best,
> Andrija
>
> On Tue, Sep 17, 2019, 08:00 Piotr Pisz <pi...@piszki.pl> wrote:
>
>> Hi all,
>>
>> I have a strange situation, we have a CephFS share mounted as
>> SharedMountPoint.
>> CS shows Disk Total as 12T (that's ok), while Disk Allocated shows
>> like 11.8T (it's not ok, disk is 50% full).
>> How can We diagnose the cause?
>>
>> Regards,
>> Piotr
>>
>>
Re: disk total vs disk allocated
Posted by Andrija Panic <an...@gmail.com>.
Wow.. . That's the community spirit - 3 replies in 8min from the original
email!
On Tue, Sep 17, 2019, 08:07 Andrija Panic <an...@gmail.com> wrote:
> Hi Piotr,
>
> Allocated should be the sum of all disks virtual size from the DB,
> "volumes" table.
>
> Can you do select sum "size" from that table, where pool_id=xxx and
> removed is NULL (on my mobile, can't remember the name of the column in the
> table, probability just "size").
>
> More importantly, I'm wondering how is this working for you? Did you have
> time to test that setup "properly"? Any feedbacks?
>
> Best,
> Andrija
>
> On Tue, Sep 17, 2019, 08:00 Piotr Pisz <pi...@piszki.pl> wrote:
>
>> Hi all,
>>
>> I have a strange situation, we have a CephFS share mounted as
>> SharedMountPoint.
>> CS shows Disk Total as 12T (that's ok), while Disk Allocated shows like
>> 11.8T (it's not ok, disk is 50% full).
>> How can We diagnose the cause?
>>
>> Regards,
>> Piotr
>>
>>
Re: disk total vs disk allocated
Posted by Andrija Panic <an...@gmail.com>.
Hi Piotr,
Allocated should be the sum of all disks virtual size from the DB,
"volumes" table.
Can you do select sum "size" from that table, where pool_id=xxx and removed
is NULL (on my mobile, can't remember the name of the column in the table,
probability just "size").
More importantly, I'm wondering how is this working for you? Did you have
time to test that setup "properly"? Any feedbacks?
Best,
Andrija
On Tue, Sep 17, 2019, 08:00 Piotr Pisz <pi...@piszki.pl> wrote:
> Hi all,
>
> I have a strange situation, we have a CephFS share mounted as
> SharedMountPoint.
> CS shows Disk Total as 12T (that's ok), while Disk Allocated shows like
> 11.8T (it's not ok, disk is 50% full).
> How can We diagnose the cause?
>
> Regards,
> Piotr
>
>
Re: disk total vs disk allocated
Posted by Ivan Kudryavtsev <ku...@bw-sw.com>.
Disk allocated != Disk used. It's for how much all volumes will span when
their thin provisioning optimization stops and they fully use the space.
вт, 17 сент. 2019 г., 13:00 Piotr Pisz <pi...@piszki.pl>:
> Hi all,
>
> I have a strange situation, we have a CephFS share mounted as
> SharedMountPoint.
> CS shows Disk Total as 12T (that's ok), while Disk Allocated shows like
> 11.8T (it's not ok, disk is 50% full).
> How can We diagnose the cause?
>
> Regards,
> Piotr
>
>