Etag support in openbmc

Etag support in openbmc

[Mohammed.Habeeb ISV]

Hi 

Do we have the Etag support in OpenBMC? I see there is a mention of ETag in 
bmcweb file redfish-core/src / error_messages.cpp, however not sure if Etag is supported.
Please let me know.

Regards

Re: Etag support in openbmc

[John Broadbent]

[-- Attachment #1: Type: text/plain, Size: 585 bytes --]

I don't think bmcweb support etags. I have never seen logic from them and
Ed made a comment about how bmcweb will have to support them in the future
here https://github.com/openbmc/webui-vue/issues/43#issue-728321615

But I defer to Ed, and Gunner.

Thank you
John Broadbent

On Mon, Oct 11, 2021 at 8:03 PM Mohammed.Habeeb ISV <
mohammed.habeeb@inventec.com> wrote:

> Hi
>
> Do we have the Etag support in OpenBMC? I see there is a mention of ETag
> in
> bmcweb file redfish-core/src / error_messages.cpp, however not sure if
> Etag is supported.
> Please let me know.
>
> Regards
>

[-- Attachment #2: Type: text/html, Size: 1034 bytes --]

Re: Etag support in openbmc

[John Broadbent]

[-- Attachment #1: Type: text/plain, Size: 854 bytes --]

ShuoX Zhang from Intel might be working on adding support, but I don't see
anything in gerrit.

https://lists.ozlabs.org/pipermail/openbmc/2021-August/027405.html

On Mon, Oct 11, 2021 at 8:37 PM John Broadbent <jebr@google.com> wrote:

> I don't think bmcweb support etags. I have never seen logic from them and
> Ed made a comment about how bmcweb will have to support them in the future
> here https://github.com/openbmc/webui-vue/issues/43#issue-728321615
>
> But I defer to Ed, and Gunner.
>
> Thank you
> John Broadbent
>
> On Mon, Oct 11, 2021 at 8:03 PM Mohammed.Habeeb ISV <
> mohammed.habeeb@inventec.com> wrote:
>
>> Hi
>>
>> Do we have the Etag support in OpenBMC? I see there is a mention of ETag
>> in
>> bmcweb file redfish-core/src / error_messages.cpp, however not sure if
>> Etag is supported.
>> Please let me know.
>>
>> Regards
>>
>

[-- Attachment #2: Type: text/html, Size: 1693 bytes --]

RE: Etag support in openbmc

[Zhang, ShuoX]

[-- Attachment #1: Type: text/plain, Size: 1218 bytes --]

Sorry, this question takes a lot of time, so it was put on hold.

From: openbmc <openbmc-bounces+shuox.zhang=intel.com@lists.ozlabs.org> On Behalf Of John Broadbent
Sent: Tuesday, October 12, 2021 11:47 AM
To: Mohammed.Habeeb ISV <mohammed.habeeb@inventec.com>
Cc: openbmc@lists.ozlabs.org
Subject: Re: Etag support in openbmc

ShuoX Zhang from Intel might be working on adding support, but I don't see anything in gerrit.

https://lists.ozlabs.org/pipermail/openbmc/2021-August/027405.html

On Mon, Oct 11, 2021 at 8:37 PM John Broadbent <jebr@google.com<mailto:jebr@google.com>> wrote:
I don't think bmcweb support etags. I have never seen logic from them and Ed made a comment about how bmcweb will have to support them in the future here https://github.com/openbmc/webui-vue/issues/43#issue-728321615

But I defer to Ed, and Gunner.

Thank you
John Broadbent

On Mon, Oct 11, 2021 at 8:03 PM Mohammed.Habeeb ISV <mohammed.habeeb@inventec.com<mailto:mohammed.habeeb@inventec.com>> wrote:
Hi

Do we have the Etag support in OpenBMC? I see there is a mention of ETag in
bmcweb file redfish-core/src / error_messages.cpp, however not sure if Etag is supported.
Please let me know.

Regards

[-- Attachment #2: Type: text/html, Size: 4473 bytes --]

Re: Etag support in openbmc

[Gunnar Mills]

> On Mon, Oct 11, 2021 at 8:37 PM John Broadbent <jebr@google.com 
> <mailto:jebr@google.com>> wrote:
> 
>     I don't think bmcweb support etags. I have never seen logic from
>     them and Ed made a comment about how bmcweb will have to support
>     them in the future here
>     https://github.com/openbmc/webui-vue/issues/43#issue-728321615
>     <https://github.com/openbmc/webui-vue/issues/43#issue-728321615>
> 

Yes, we don't currently support etag. As mentioned in
https://lists.ozlabs.org/pipermail/openbmc/2021-September/027406.html
if you plan to add, we will want to make sure we do this correctly.

Thanks,
Gunnar

Re: Etag support in openbmc

[Ed Tanous]

On Tue, Oct 12, 2021 at 10:31 AM Gunnar Mills <gmills@linux.vnet.ibm.com> wrote:
>
> > On Mon, Oct 11, 2021 at 8:37 PM John Broadbent <jebr@google.com
> > <mailto:jebr@google.com>> wrote:
> >
> >     I don't think bmcweb support etags. I have never seen logic from
> >     them and Ed made a comment about how bmcweb will have to support
> >     them in the future here
> >     https://github.com/openbmc/webui-vue/issues/43#issue-728321615
> >     <https://github.com/openbmc/webui-vue/issues/43#issue-728321615>
> >
>
> Yes, we don't currently support etag. As mentioned in
> https://lists.ozlabs.org/pipermail/openbmc/2021-September/027406.html
> if you plan to add, we will want to make sure we do this correctly.


+1.  Just some additional background, at one point in its history
bmcweb supported etag for the static resources (in fact, there's still
a disabled unit test for it in the repo), but it got removed at the
same time when it went from the built-in angularjs bmcweb UI, and
moved to phosphor-webui, because the etags were being generated from
the hashing procedure, which no longer happened in the bmcweb build,
so the data was harder to get.  In theory we can parse the filename,
look for the webpack-like value, and use that as the hash value for
the etag again, but in practice, it didn't really make things faster
in most UI cases, so I never really cared to add it back.

I realize you're likely talking about caching of Redfish resources,
which is a whole different (and much more complex) thing to implement.

>
> Thanks,
> Gunnar
>

Re: Etag support in openbmc

[Ed Tanous]

We previously supported etag for static resources, hence the
references to them.  They were removed when I moved phosphor-webui to
webpack builds, and it was always on my list to add them back, using
webpacks hashing mechanisms, but I never got a chance.

In practice, they didn't provide that much utility because most browsers:
1. only cache if you have a valid SSL certificate.
2. Have a very limited cache size.
3. Don't support cross site etag caches.

So in practice, the cache didn't provide any benefit in a majority of cases.

In terms of supporting etags on non-static resources, that's quite
complex, given that cache invalidation is difficult, and likely
deserves its own design document.

-Ed

On Mon, Oct 11, 2021 at 8:04 PM Mohammed.Habeeb ISV
<mohammed.habeeb@inventec.com> wrote:
>
> Hi
>
> Do we have the Etag support in OpenBMC? I see there is a mention of ETag in
> bmcweb file redfish-core/src / error_messages.cpp, however not sure if Etag is supported.
> Please let me know.
>
> Regards