zheap storage_engine parameter, shouldn't this raise an error?

Started by Daniel Westermannabout 7 years ago3 messages
#1Daniel Westermann
daniel.westermann@dbi-services.com
1 attachment(s)

Hi,

I believe this should raise an error or at least a warning?

postgres=# alter system set storage_engine = 'zheap';
ALTER SYSTEM
postgres=# alter system set storage_engine = 'zheap2';
ALTER SYSTEM
postgres=# alter system set storage_engine = 'zheap3';
ALTER SYSTEM
postgres=#

Regards
Daniel

[cid:9bfad55c-fe48-4c76-8ec8-8b344f9b4094]
Daniel Westermann | Delivery Manager & Open Infrastructure Technology Leader
Phone: +41 32 422 96 00 | Mobile: +41 79 927 24 46 | Fax: +41 32 422 96 15
dbi services, Voltastrasse 104 | CH-4056 Basel
daniel.westermann@dbi-services.com
www.dbi-services.com

Attachments:

Outlook-lp0cbmpn.pngimage/png; name=Outlook-lp0cbmpn.pngDownload
#2Amit Kapila
amit.kapila16@gmail.com
In reply to: Daniel Westermann (#1)
1 attachment(s)
Re: zheap storage_engine parameter, shouldn't this raise an error?

On Sat, Nov 3, 2018 at 3:13 PM Daniel Westermann <
daniel.westermann@dbi-services.com> wrote:

Hi,

I believe this should raise an error or at least a warning?

postgres=# alter system set storage_engine = 'zheap';
ALTER SYSTEM
postgres=# alter system set storage_engine = 'zheap2';
ALTER SYSTEM
postgres=# alter system set storage_engine = 'zheap3';
ALTER SYSTEM
postgres=#

Sure, if we want we can raise the error or warning for this, but this is a
parameter mainly to test zheap with existing set of regression tests. I am
not sure if we want to keep it or even if we want to have any such
parameter for testing purpose in what form it will be present. So, adding
more checks at this stage around this parameter doesn't seem advisable to
me.

--
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com

Attachments:

Outlook-lp0cbmpn.pngimage/png; name=Outlook-lp0cbmpn.pngDownload
#3Amit Kapila
amit.kapila16@gmail.com
In reply to: Amit Kapila (#2)
Re: zheap storage_engine parameter, shouldn't this raise an error?

On Sat, Nov 3, 2018 at 3:21 PM Amit Kapila <amit.kapila16@gmail.com> wrote:

On Sat, Nov 3, 2018 at 3:13 PM Daniel Westermann <daniel.westermann@dbi-services.com> wrote:

Hi,

I believe this should raise an error or at least a warning?

Sure, if we want we can raise the error or warning for this, but this is a parameter mainly to test zheap with existing set of regression tests. I am not sure if we want to keep it or even if we want to have any such parameter for testing purpose in what form it will be present. So, adding more checks at this stage around this parameter doesn't seem advisable to me.

Instead of starting new threads for each report, it might be better to
report it on the zheap thread [1]/messages/by-id/CAA4eK1+YtM5vxzSM2NZm+pC37MCwyvtkmJrO_yRBQeZDp9Wa2w@mail.gmail.com.

[1]: /messages/by-id/CAA4eK1+YtM5vxzSM2NZm+pC37MCwyvtkmJrO_yRBQeZDp9Wa2w@mail.gmail.com

--
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com