Donal had a good comment on the wiki page, and I replied. Maybe that would help
answer your question of ID:
https://cwiki.apache.org/confluence/display/GEODE/Make+Cluster+Management+Service(CMS)+Thread+Safe
From: Anilkumar Gingade
Sent: Thursday, May 28, 2020
Yes, the DLock machinery handles (has option) dlock grantor departure...
As I understand, right now we have dlock at config persistence layer, but this
does not guarantee preserving the order in which the config changes are
applied. E.g.: A create region command followed by destroy could be
+1
On Thu, May 28, 2020 at 11:49 AM Ernie Burghardt
wrote:
> +1
>
> On 5/27/20, 1:35 PM, "Bruce Schuchardt" wrote:
>
> This ticket has two PRs. One passed all normal CI runs but then we
> hit a faulty test that failed on a Windows machine. There’s a new PR that
> fixes that test & has
+1
On 5/27/20, 1:35 PM, "Bruce Schuchardt" wrote:
This ticket has two PRs. One passed all normal CI runs but then we hit a
faulty test that failed on a Windows machine. There’s a new PR that fixes that
test & has been merged.
The PRs fixe endpoint verification problems in servers
Simultaneous updates to configurations are already protected by a different
dlock, so I assume they can be made safely.
Typically a CMS operation involves two parts:
1) updates to the servers to "realize" the configuration
2) updates to the configurations to "persist" it.
The purpose of the
I'm going to spend a release-manager +1 to put this proposal over the top.
Please merge this fix into support/1.13, Bruce.
Thanks,
Dave
On Thu, May 28, 2020 at 7:52 AM Udo Kohlmeyer wrote:
> +1
> On May 27, 2020, 1:35 PM -0700, Bruce Schuchardt ,
> wrote:
> This ticket has two PRs. One passed
Hi Geode Community,
Juan Ramos has recently published a new article on Spring Security and Apache
Geode. You can read it here at
https://medium.com/@jujoramos/spring-security-geode-4670faff47a0
Thank you Juan for this article.
Regards
Nabarun Nag
Sadly not, unless it was incredibly convoluted and complex. There are plenty of
String literal "/" still in the codebase, in URIs/URLs, filepaths and log
output (for example "Updated 5/6 values") so it's not really possible to
determine if the presence of a "/" is "correct" without looking at
Is there any way to enforce that with some kind of LGTM or spotless rule?
On 5/28/20, 12:46 PM, "Donal Evans" wrote:
I'm happy to say that as of about 5 minutes ago, there are no uses of
hardcoded "/" in region paths/names in the geode codebase, as all of them have
been replaced by the
I think the first question to answer is: can simultaneous updates to
configuration be made safely? Or what is the critical section of code that
needs to be protected?
Another thing to consider with dlocks is what happens in the failure case when
the lock is not properly released. Does it
Thanks for the suggestion, Dave. I'll be sure to add something soon.
From: Dave Barnes
Sent: Thursday, May 28, 2020 10:32 AM
To: dev@geode.apache.org
Subject: Re: No more hardcoded region separators!
Excellent, Donal!
If you have not already done so, please
The proposal is proposing using ONE single dlock to synchronize all CMS CRUD
operations, that means, in a given time, only one CRUD operation in CMS is
allowed in the entire cluster, this seems a bit too harsh.
Another way is to use a dlock per ID to only synchronize CRUD operation on the
same
Excellent, Donal!
If you have not already done so, please consider documenting the practice
you're advocating in a place where all community contributors have a chance
of seeing it. Maybe
https://cwiki.apache.org/confluence/display/GEODE/How+to+Contribute?
On Thu, May 28, 2020 at 9:46 AM Donal
I'm happy to say that as of about 5 minutes ago, there are no uses of hardcoded
"/" in region paths/names in the geode codebase, as all of them have been
replaced by the Region.SEPARATOR constant (with the exception of a few
occurrences in the geode-management module, which while not having an
Hi, Geode Community,
Currently, the CMS CRUD operations are not thread safe, if one call tries to
create a region, and another call tries to delete the same region, if timing is
off, we could end up with inconsistent state (what's in cluster config and
what's actually on the server). So we
Please review my PR if you have time:
https://github.com/apache/geode/pull/5162
This PR fixes flakiness in ShutdownCommandOverHttpDUnitTest by adding an
IgnoredException.
+1
On May 27, 2020, 1:35 PM -0700, Bruce Schuchardt , wrote:
This ticket has two PRs. One passed all normal CI runs but then we hit a faulty
test that failed on a Windows machine. There’s a new PR that fixes that test &
has been merged.
The PRs fixe endpoint verification problems in servers and
+1
From: Bruce Schuchardt
Sent: May 27, 2020 16:35
To: dev@geode.apache.org
Subject: [PROPOASAL] backport GEODE-8144
This ticket has two PRs. One passed all normal CI runs but then we hit a
faulty test that failed on a Windows machine. There’s a new PR that
18 matches
Mail list logo
