PERTANYAANNYA
Why does the same outage keep recurring?
Gunakan stiker Anda untuk hal yang paling penting. Gagasan yang disertai bukti bisa menempati kursi khusus.
ditutup · kembali ke tahap Tindakan
Melihat sebagai tamu. Masukkan nama pengguna untuk bergabung dan ikut serta.
Jatah stiker Anda
Satu per gagasan, total 5. Minimal satu suara.
5 tempat disediakan untuk gagasan yang terlewat — calonkan satu dengan tautan, dan gagasan itu berlaku setelah 2 orang setuju sumbernya relevan.
mengapa
Saat tahap ditutup, satu calon masuk lewat undian dan sisanya berdasarkan jumlah pencalonan, paling banyak satu per kelas pemangku kepentingan. Model hanya memeriksa apakah halamannya membahas pokok yang sama — tidak pernah memeriksa apakah halaman itu membuktikan klaimnya. Penilaian itu milik ruang, dan pencalonan yang tidak didukung orang lain tidak mendapat kursi.
One engineer knows how the payment path actually works
Scope: the services this team runs, not the whole platform.
Belum ada yang berbicara.
KLASTER 1
The on-call rota has the same two people on it most weeks
7 suaradiskusikan
The same three services cause most of the pages
14 suaradiskusikan
KLASTER 2
Post-mortem actions are written up and then never scheduled
4 suaradiskusikan
Feature flags are never cleaned up, so nobody knows which paths are live
Meaning the repeat outages, not the one-off hardware failure in March.
7 suaradiskusikan
KLASTER 3
Migrations are run by hand from a laptop
2 suaradiskusikan
Config lives in five places and drifts between them
I mean the pattern over the last two quarters, not a single incident.
3 suaradiskusikan
KLASTER 4
Reliability has no budget line of its own
To be precise: this is about the process, not the person on call that night.
11 suaradiskusikan
Every team has its own logging format
3 suaradiskusikan
KLASTER 5
Customers report outages before monitoring does
I mean the pattern over the last two quarters, not a single incident.
10 suaradiskusikan
Alerts fire so often that on-call mutes them
I mean the pattern over the last two quarters, not a single incident.
6 suaradiskusikan
KLASTER 6
The service map in the wiki is two years old
5 suaradiskusikan
Runbooks are out of date within a month of being written
To be precise: this is about the process, not the person on call that night.
2 suaradiskusi (1)
TANPA KLASTER
Tests that fail intermittently are retried until green
10 suaradiskusikan
Leadership asks for a root cause within the hour, so the first plausible one wins
I mean the pattern over the last two quarters, not a single incident.
1 suaradiskusikan
Success is measured in features shipped per quarter
1 suaradiskusikan
Dependencies are upgraded only when something breaks
Meaning the repeat outages, not the one-off hardware failure in March.
0 suaradiskusikan
Deploys go out on Friday afternoons because that is when the sprint ends
2 suaradiskusikan
Rollbacks take longer than the outage they are meant to end
To be precise: this is about the process, not the person on call that night.
3 suaradiskusikan
Post-mortems assign actions to people who were not in the room
Scope: the services this team runs, not the whole platform.
8 suaradiskusi (1)
Feature work always outranks reliability work in planning
Scope: the services this team runs, not the whole platform.
4 suaradiskusikan
There is no time set aside after an incident, so the write-up is done at midnight
2 suaradiskusikan
Incidents are declared late because declaring one feels like blame
1 suaradiskusikan
Load tests were last run before the customer base doubled
1 suaradiskusikan
One engineer knows how the payment path actually works
Scope: the services this team runs, not the whole platform.
5 suara
↳ dicalonkan untuk kursi: https://example.org/ops-incident-review.pdf could not read it (HTTPError) didukung 1 dari 2
disorot ↑ · tutup ✕Error budgets exist on a slide and nowhere else
8 suaradiskusikan
The staging environment does not resemble production
7 suaradiskusikan
The incident channel fills with people asking for status instead of giving it
3 suaradiskusikan
Nobody owns the shared database schema
To be precise: this is about the process, not the person on call that night.
2 suaradiskusikan
Retries are unbounded, so a slow dependency becomes a flood
6 suaradiskusikan
Capacity is added after the outage, never before
0 suaradiskusikan