# Tell me (if you can) why I shouldn't worry about the dashboards
**URL:** https://discourse.vtk.org/t/tell-me-if-you-can-why-i-shouldnt-worry-about-the-dashboards/5610
**Category:** Development
**Created:** [April 22, 2021, 12:15pm UTC](https://discourse.vtk.org/t/tell-me-if-you-can-why-i-shouldnt-worry-about-the-dashboards/5610 "2021-04-22T12:15:42Z")
**Posts on this page:** 10
**Page:** 1
### Author:  [@will.schroeder](https://discourse.vtk.org/u/will.schroeder)
#### Post date: [April 22, 2021, 12:15pm UTC](https://discourse.vtk.org/t/tell-me-if-you-can-why-i-shouldnt-worry-about-the-dashboards/5610/1 "2021-04-22T12:15:42Z")
The VTK dashboards are scary. I know a lot is going on, but the current state worries me. Should we go into rapid response mode and fix this ? I see several sloppy warnings/errors that have been around for a while, at a minimum we can clean that junk up.
---
### Author:  [@ben.boeckel](https://discourse.vtk.org/u/ben.boeckel)
#### Post date: [April 22, 2021, 12:39pm UTC](https://discourse.vtk.org/t/tell-me-if-you-can-why-i-shouldnt-worry-about-the-dashboards/5610/2 "2021-04-22T12:39:08Z")
The current test failures are from [this MR](https://gitlab.kitware.com/vtk/vtk/-/merge_requests/7844) and is being reverted in [this MR](https://gitlab.kitware.com/vtk/vtk/-/merge_requests/7873). The build failures from out of space…I’m on that now; it seems one of the CI machines has just a 200G disk which shouldn’t be a thing (we aim for 1TB at least).
The transition to CI is currently disruptive, but I’ve been trying to keep `master` clean. Developers not running CI or assuming problems aren’t theirs seem to be the main issue right now.
---
### Author:  [@will.schroeder](https://discourse.vtk.org/u/will.schroeder)
#### Post date: [April 22, 2021, 1:09pm UTC](https://discourse.vtk.org/t/tell-me-if-you-can-why-i-shouldnt-worry-about-the-dashboards/5610/3 "2021-04-22T13:09:35Z")
Thanks, I will address some slop issues today
---
### Author:  [@ben.boeckel](https://discourse.vtk.org/u/ben.boeckel)
#### Post date: [April 22, 2021, 1:22pm UTC](https://discourse.vtk.org/t/tell-me-if-you-can-why-i-shouldnt-worry-about-the-dashboards/5610/4 "2021-04-22T13:22:27Z")
[This MR](https://gitlab.kitware.com/vtk/vtk/-/merge_requests/7869) fixes a warning in the Imprint filter if you’re looking at that one.
---
### Author:  [@will.schroeder](https://discourse.vtk.org/u/will.schroeder)
#### Post date: [April 22, 2021, 1:26pm UTC](https://discourse.vtk.org/t/tell-me-if-you-can-why-i-shouldnt-worry-about-the-dashboards/5610/5 "2021-04-22T13:26:18Z")
Thanks! Ben I notice builds like [https://open.cdash.org/build/7182539](https://open.cdash.org/build/7182539) are recommending replacing vector\<\>::push\_back with emplace\_back. (I’ve been doing this for a while now anyway in new classes.) Do you think it’s a good idea to do this wherever push\_back is found?
---
### Author:  [@ben.boeckel](https://discourse.vtk.org/u/ben.boeckel)
#### Post date: [April 22, 2021, 2:19pm UTC](https://discourse.vtk.org/t/tell-me-if-you-can-why-i-shouldnt-worry-about-the-dashboards/5610/6 "2021-04-22T14:19:39Z")
No, it shouldn’t be a knee-jerk reaction. It should be done when a move is possible and better. For example:
```auto
std::vector v;
// always better as v.emplace_back(/*ctor args*/);
v.push_back(Type(/*ctor args*/));
// Only better if `t` isn't used after the insertion.
Type t;
v.push_back(t);
// Always better.
v.push_back(std::move(t)); // `t` is "dead" after this anyways
```
For more details, [this blog post](https://quuxplusone.github.io/blog/2021/03/03/push-back-emplace-back/) might help.
---
### Author:  [@seanm](https://discourse.vtk.org/u/seanm)
#### Post date: [April 22, 2021, 4:22pm UTC](https://discourse.vtk.org/t/tell-me-if-you-can-why-i-shouldnt-worry-about-the-dashboards/5610/7 "2021-04-22T16:22:00Z")
Some weeks ago, I had my bots (the Rogue Research ones) down to zero errors/warnings but just days later there were new ones. I have this suspicion that no one looks at cdash anymore and only looks at gitlab. ☹
---
### Author:  [@ben.boeckel](https://discourse.vtk.org/u/ben.boeckel)
#### Post date: [April 22, 2021, 5:21pm UTC](https://discourse.vtk.org/t/tell-me-if-you-can-why-i-shouldnt-worry-about-the-dashboards/5610/8 "2021-04-22T17:21:12Z")
We can add warning flags to the CI bots to match the Rogue machines. The thing is that waiting for CI on MRs, then _also_ checking CDash once it has been merged for followups (without a way to verify warning fixes) is not any easier.
Right now, there are a bunch of VTK-m deprecation warnings that fire on the submodule I just updated to, but it’s a similar thing as other cases: things just need to move forward or I’m stalled out waiting for feedback on what to do which isn’t viable because everybody else is busy too.
---
### Author:  [@ben.boeckel](https://discourse.vtk.org/u/ben.boeckel)
#### Post date: [April 22, 2021, 7:03pm UTC](https://discourse.vtk.org/t/tell-me-if-you-can-why-i-shouldnt-worry-about-the-dashboards/5610/9 "2021-04-22T19:03:59Z")
VTK-m warnings (from non-Fides code) should be resolved in [https://gitlab.kitware.com/vtk/vtk/-/merge\_requests/7879](https://gitlab.kitware.com/vtk/vtk/-/merge_requests/7879)
---
### Author:  [@dgobbi](https://discourse.vtk.org/u/dgobbi)
#### Post date: [April 22, 2021, 8:30pm UTC](https://discourse.vtk.org/t/tell-me-if-you-can-why-i-shouldnt-worry-about-the-dashboards/5610/10 "2021-04-22T20:30:23Z")
A fix for the DICOMParser warning is [!7882](https://gitlab.kitware.com/vtk/vtk/-/merge_requests/7882).