Processing Kubectl and Helm output¶
The kubectl
and helm
commands produce tabular-looking output, which Miller can parse -- however,
there's a bit of whitespace-handling to be dealt with first.
Whitespace structure of kubectl¶
The output of kubectl
looks tabular -- so PPRINT format is perhaps a good choice.
$ kubectl -n my-namespace get pods | head NAME READY STATUS RESTARTS AGE app-5mjwm4-274754k8468 0/1 Completed 0 6m51s app-5mjwm4-274754vdfnf 0/1 Completed 0 6m50s app-5mjwm4-0 1/1 Running 0 6h8m app-5mjwm4-27475nt9cc 0/1 Completed 0 6m53s app-5mjwm4-27474454-dc7wq 0/1 Error 0 16h app-5mjwm4-27475416-tv2ff 0/1 Completed 0 56s app-5mjwm4-2747541t7lgk 0/1 Completed 0 115s app-5mjwm4-27475245-7sg9r 0/1 Completed 0 171m app-5mjwm4-27475410-k4gcr 0/1 Completed 0 6m52s
We can verify this using kubectl -n my-namespace get pods | vim -
, then :set list
within vim
;
or, perhaps piping the output to cat -t
, or bat -A -- in any
case what looks like whitespace is really all space characters.
To double-check, it's helpful to run tabular-looking output through a format-converter, and make sure the column headers are being correctly identified as keys, and the remaining lines are being correctly identified as values:
$ kubectl -n my-namespace get pods | mlr --ipprint --ojson head -n 1 [ { "NAME": "app-5mjwm4-274754k8468", "READY": "0/1", "STATUS": "Completed", "RESTARTS": 0, "AGE": "14m" } ]
Sorting/filtering¶
Suppose we want to sort the information for non-completed pods by age. We can use
dhms2sec to turn the AGE
into something sortable.
$ kubectl -n service-xyz get pods \ | mlr --pprint \ filter '$STATUS != "Completed"' \ then put '$AGESEC = dhms2sec($AGE)' \ then sort -n AGESEC NAME READY STATUS RESTARTS AGE AGESEC app1-1500-5mjwm4-0 1/1 Running 0 6h22m 22920 app1-1624-6dh711-0 1/1 Running 0 6h27m 23220 app1-1500-pqb9b4-0 1/1 Running 0 6h30m 23400 app1-gbwuwi-2747495lbtzg 0/1 Error 0 7h59m 28740 app1-gbwuwi-0 1/1 Running 0 8h 28800 app1-gbwuwi-27474955r8gq 0/1 Error 0 8h 28800 app1-gbwuwi-27474956rps8 0/1 Error 0 8h 28800 app1-gbwuwi-2747495q7fnz 0/1 Error 0 8h 28800 app1-gbwuwi-2747495vnxgn 0/1 Error 0 8h 28800 app1-gbwuwi-674ddcfd89-2jt64 2/2 Running 0 8h 28800 app3-5c79574b69-8njgr 2/2 Running 0 9h 32400 app3-5c79574b69-np2qj 2/2 Running 0 9h 32400 app3-a56i7c-0 1/1 Running 0 13h 46800 app3-a56i7c-587dfc99cf-zrr4t 2/2 Running 0 13h 46800 app2-1500-pqb9b4-274746pfbfd 0/1 Error 0 13h 46800 app2-1500-pqb9b4-274746jtz8t 0/1 Error 0 13h 46800 app2-1500-pqb9b4-274746pmmhq 0/1 Error 0 13h 46800 app2-1500-pqb9b4-27474624h8fp 0/1 Error 0 13h 46800 app2-1500-pqb9b4-2747462d8n96 0/1 Error 0 13h 46800 app2-1500-pqb9b4-2747462xnmcf 0/1 Error 0 13h 46800 app2-1500-pqb9b4-27474630-95668 0/1 Error 0 13h 46800 app1-1500-pqb9b4-sr5vd 2/2 Running 0 13h 46800 app1-1500-5mjwm4-27474454-dc7wq 0/1 Error 0 16h 57600 app1-1500-5mjwm4-667c6fc66d-b97m9 2/2 Running 0 16h 57600 app1-1624-6dh711-2747435h42j 0/1 Error 0 17h 61200 app1-1624-6dh711-27474370-ph25r 0/1 Error 0 17h 61200 app1-1624-6dh711-74fb5cf9d6-cl5tq 2/2 Running 0 17h 61200
Whitespace structure of helm list¶
The output of helm list
is a bit fussier. Here it's already clear that something's amiss, since not everything lines up:
$ helm list NAME NAMESPACE REVISION UPDATED STATUS CHART APP VERSION appdev-an-sc-1500-5mjwm4 service-xyz 1 2022-03-28 11:33:05.389975262 +0000 UTC deployed appdev-cloud-test-7.1.12 appdev-exyzv-load-a56i7c service-xyz 1 2022-03-28 14:45:35.44317196 +0000 UTC deployed appdev-cloud-test-7.1.12 appdev-sa-sc-1500-pqb9b4 service-xyz 1 2022-03-28 14:24:33.978580048 +0000 UTC deployed appdev-cloud-test-7.1.12 appdev-sa-sc-1624-6dh711 service-xyz 1 2022-03-28 10:09:05.966332699 +0000 UTC deployed appdev-cloud-test-7.1.12 appdev-wertzxyffa-gbwuwi service-xyz 1 2022-03-28 19:47:34.96763583 +0000 UTC deployed appdev-cloud-test-7.1.12 staging service-xyz 797 2022-03-28 18:39:34.005120936 +0000 UTC deployed appdev-cloud-test-7.1.12
This isn't likely to be PPRINT format, as we soon see. Also note that the space before +0000
is an issue.
$ helm list | mlr --ipprint --ojson cat mlr : mlr: CSV header/data length mismatch 7 != 5 at filename (stdin) line 2.
Running through bat -A
or cat -t
shows an issue. Namely, the Helm authors are mixing tabs and spaces -- cat -t
shows tabs as ^I
:
$ helm list | cat -t NAME ^INAMESPACE ^IREVISION^IUPDATED ^ISTATUS ^ICHART ^IAPP VERSION appdev-an-sc-1500-5mjwm4^Iservice-xyz^I1 ^I2022-03-28 11:33:05.389975262 +0000 UTC^Ideployed^Iappdev-cloud-test-7.1.12^I appdev-exyzv-load-a56i7c^Iservice-xyz^I1 ^I2022-03-28 14:45:35.44317196 +0000 UTC ^Ideployed^Iappdev-cloud-test-7.1.12^I appdev-sa-sc-1500-pqb9b4^Iservice-xyz^I1 ^I2022-03-28 14:24:33.978580048 +0000 UTC^Ideployed^Iappdev-cloud-test-7.1.12^I appdev-sa-sc-1624-6dh711^Iservice-xyz^I1 ^I2022-03-28 10:09:05.966332699 +0000 UTC^Ideployed^Iappdev-cloud-test-7.1.12^I appdev-wertzxyffa-gbwuwi^Iservice-xyz^I1 ^I2022-03-28 19:47:34.96763583 +0000 UTC ^Ideployed^Iappdev-cloud-test-7.1.12^I staging ^Iservice-xyz^I797 ^I2022-03-28 18:39:34.005120936 +0000 UTC^Ideployed^Iappdev-cloud-test-7.1.12^I
This mix of tabs and spaces, while not PPRINT, also isn't quite TSV either. As above, it's helpful to run tabular-looking data through a format-converter to see how it's structured:
$ helm list | mlr --itsv --ojson head -n 1 [ { "NAME ": "appdev-an-sc-1500-5mjwm4", "NAMESPACE ": "service-xyz", "REVISION": "1 ", "UPDATED ": "2022-03-28 11:33:05.389975262 +0000 UTC", "STATUS ": "deployed", "CHART ": "appdev-cloud-test-7.1.12", "APP VERSION": " " } ]
A solution here is Miller's clean-whitespace verb:
$ helm list | mlr --itsv --ojson clean-whitespace then head -n 1 [ { "NAME": "appdev-an-sc-1500-5mjwm4", "NAMESPACE": "service-xyz", "REVISION": "1 ", "UPDATED": "2022-03-28 11:33:05.389975262 +0000 UTC", "STATUS ": "deployed", "CHART": "appdev-cloud-test-7.1.12", "APP VERSION": "" } ]
Now we have the keys and values correctly identified within the tabular-looking data.
Sorting/filtering¶
To find oldest items, it would suffice to sort by the UPDATED
column, as that sorts lexically.
However, let's parse the timestamps and compute their ages from the present:
$ helm list \ | mlr --itsv --opprint clean-whitespace \ then put '$AGESEC = int(systime() - strptime($UPDATED, "%Y-%m-%d %H:%M:%S.%f +0000 UTC"))' \ then sort -n AGESEC \ then cut -x -f 'APP VERSION,UPDATED' NAME NAMESPACE REVISION STATUS CHART AGESEC appdev-sa-sc-1624-6dh711 service-xyz 1 deployed appdev-cloud-test-7.1.12 30874 appdev-an-sc-1500-5mjwm4 service-xyz 797 deployed appdev-cloud-test-7.1.12 34955 appdev-sa-sc-1500-pqb9b4 service-xyz 1 deployed appdev-cloud-test-7.1.12 48993 appdev-xxyzv-load-a56i7c service-xyz 1 deployed appdev-cloud-test-7.1.12 50255 staging service-xyz 1 deployed appdev-cloud-test-7.1.12 60543 appdev-wertzxyffa-gbwuwi service-xyz 1 deployed appdev-cloud-test-7.1.12 65583
Extracting fields to be acted on¶
Switching to NIDX format lets us extract fields and pass them onto other commands -- e.g. helm uninstall
.
We just need to switch the output format to --onidx
, then cut out the NAME
field. (Maybe add then filter $AGESEC > 86400
or somesuch.)
$ helm list \ | mlr --itsv --onidx clean-whitespace \ then put '$UPDATED = ssub($UPDATED, " +0000 UTC", "")' \ then put '$AGESEC = int(systime() - strptime($UPDATED, "%Y-%m-%d %H:%M:%S.%f"))' \ then sort -n AGESEC \ then cut -f NAME \ | tee names.txt appdev-sa-sc-1624-6dh711 appdev-an-sc-1500-5mjwm4 appdev-sa-sc-1500-pqb9b4 appdev-xxyzv-load-a56i7c staging appdev-wertzxyffa-gbwuwi
Then
$ for name in $(cat names.txt); do helm uninstall $name; done