control.sh: a bug or feature?

[Max Shonichev]

Hello, Igniters!

AI 2.6 has added `--cache idle_verify command` to control.sh.

Upcoming changes in 2.7 would add some optional arguments, specifically `--dump`. Also, all `control.sh` commands, including `--cache`, have optional connection arguments like, e.g. `--host`, `--port`.

During testing I found that order of optional arguments is crucial to success.

In other words, this command works as expected:

```
bin/control.sh --host 172.25.1.14 --cache idle_verify --dump
```

while this command miserably fails:
```
bin/control.sh --cache idle_verify --host 172.25.1.14 --dump
```

I consider this a behaviour unexpected to user and filed an issue to improve

https://ggsystems.atlassian.net/browse/IGN-12325

Later on, during personal discussion with developer, we argue whether this is actually a bug from user experience point of view or not, so I would like to ask your help voting for whether this should be fixed or not.

---
Thanks in advance,
Max

[Vladimir Ozerov]

I think this is definitely a bug. From our experience with millions of
other utilities we know, that relative order of parameters with "-" or "--"
prefixes against each other doesn't matter. This is what users will expect
from our utility too.

On Thu, Nov 15, 2018 at 11:05 AM mshonich <[hidden email]> wrote:

> Hello, Igniters!
>
> AI 2.6 has added `--cache idle_verify command` to control.sh.
>
> Upcoming changes in 2.7 would add some optional arguments, specifically
> `--dump`. Also, all `control.sh` commands, including `--cache`, have
> optional connection arguments like, e.g. `--host`, `--port`.
>
> During testing I found that order of optional arguments is crucial to
> success.
>
> In other words, this command works as expected:
>
> ```
> bin/control.sh --host 172.25.1.14 --cache idle_verify --dump
> ```
>
> while this command miserably fails:
> ```
> bin/control.sh --cache idle_verify --host 172.25.1.14 --dump
> ```
>
> I consider this a behaviour unexpected to user and filed an issue to
> improve
>
> https://ggsystems.atlassian.net/browse/IGN-12325
>
> Later on, during personal discussion with developer, we argue whether this
> is actually a bug from user experience point of view or not, so I would
> like to ask your help voting for whether this should be fixed or not.
>
> ---
> Thanks in advance,
> Max
>

[yzhdanov]

Max, correct link to ticket is
https://issues.apache.org/jira/browse/IGNITE-10258

I would agree with you and Vladimir that parameters in mentioned case may
appear in any order.

--Yakov