text stringlengths 7 369 |
|---|
[2251.38 --> 2253.24] well I think the |
[2253.24 --> 2254.04] obvious one for me is |
[2254.04 --> 2254.68] something like health |
[2254.68 --> 2256.52] checks dot IO so |
[2256.52 --> 2258.12] what this does is when |
[2258.12 --> 2259.80] you run a cron job you |
[2259.80 --> 2261.12] append to the end of |
[2261.12 --> 2263.12] your script or your |
[2263.12 --> 2264.06] bash command that you're |
[2264.06 --> 2266.00] putting into cron a |
[2266.00 --> 2268.24] little curl with a uuid |
[2268.24 --> 2270.26] in it so curl htps |
[2270.26 --> 2273.32] url slash and then you |
[2273.32 --> 2274.62] put your little uuid in |
[2274.62 --> 2276.18] there and what that does |
[2276.18 --> 2277.18] is it pings the health |
[2277.18 --> 2278.40] check server and says |
[2278.40 --> 2280.94] hey this this job just |
[2280.94 --> 2282.08] ran and what you can then |
[2282.08 --> 2283.54] do is configure timeouts |
[2283.54 --> 2285.06] to say right I want |
[2285.06 --> 2286.06] this job to run every |
[2286.06 --> 2286.92] day and the grace |
[2286.92 --> 2288.06] period is six hours |
[2288.06 --> 2289.56] because sometimes a |
[2289.56 --> 2291.46] backup can take six |
[2291.46 --> 2292.92] hours to complete so |
[2292.92 --> 2295.16] it reduces the noise |
[2295.16 --> 2295.64] and the false |
[2295.64 --> 2296.62] positives you might get |
[2296.62 --> 2298.20] there sometimes but the |
[2298.20 --> 2299.18] other really cool thing |
[2299.18 --> 2300.46] you can do and some of |
[2300.46 --> 2301.28] the guys in our discord |
[2301.28 --> 2302.62] have been experimenting |
[2302.62 --> 2303.78] with health checks a lot |
[2303.78 --> 2304.74] of which there is a |
[2304.74 --> 2305.90] self-hosted version by the |
[2305.90 --> 2306.78] way as well don't be |
[2306.78 --> 2307.70] confused and think it's |
[2307.70 --> 2308.84] just a service it's not |
[2308.84 --> 2311.36] what you can also do is |
[2311.36 --> 2313.38] put a slash start at the |
[2313.38 --> 2314.38] beginning of your script |
[2314.38 --> 2315.80] to that same url with the |
[2315.80 --> 2317.54] uuid as well and then |
[2317.54 --> 2319.60] you can monitor how long |
[2319.60 --> 2321.86] specific jobs take so you |
[2321.86 --> 2323.90] can look and say oh my |
[2323.90 --> 2326.00] backups on average take two |
[2326.00 --> 2327.88] minutes 30 seconds this one |
[2327.88 --> 2329.50] took four hours why is that |
[2329.50 --> 2330.60] and then you can go and |
[2330.60 --> 2332.32] look at the logs because you |
[2332.32 --> 2333.94] can append logs to health |
[2333.94 --> 2335.04] checks as well if you didn't |
[2335.04 --> 2337.78] know that you can use curl to |
[2337.78 --> 2339.84] do that as well so I mean |
[2339.84 --> 2341.52] for me yes you could go and |
[2341.52 --> 2342.98] you know reinvent the wheel |
[2342.98 --> 2344.02] and have something that |
[2344.02 --> 2345.54] does log passing and alert |
[2345.54 --> 2346.90] manager with prometheus and |
[2346.90 --> 2349.62] get as complex as you like |
[2349.62 --> 2351.60] but really for me keep it |
[2351.60 --> 2354.04] simple because if you don't |
[2354.04 --> 2356.14] it'll just you just won't |
[2356.14 --> 2357.30] keep on top of it if it's a |
[2357.30 --> 2360.16] personal system yep yep and |
[2360.16 --> 2361.42] if it's going to send you an |
[2361.42 --> 2363.06] alert make sure it's sending |
[2363.06 --> 2363.94] you an alert somewhere you're |
[2363.94 --> 2365.66] actually checking and that |
[2365.66 --> 2366.70] you won't just ignore that |
[2366.70 --> 2367.80] doesn't just end up in some |
[2367.80 --> 2369.76] folder on Gmail or something |
[2369.76 --> 2371.44] like that all right well our |
[2371.44 --> 2373.46] next email comes in from |
[2373.46 --> 2375.92] Norm Norm writes in hey |
[2375.92 --> 2376.98] guys love the show self |
[2376.98 --> 2377.90] hosted alone is making me |
[2377.90 --> 2378.58] want to purchase a |
[2378.58 --> 2379.56] membership to support you |
[2379.56 --> 2380.88] thank you Norm self |
[2380.88 --> 2382.94] hosted dot show slash sre for |
[2382.94 --> 2384.38] this show or jupiter dot |
[2384.38 --> 2385.50] party for all the shows |
[2385.50 --> 2387.12] independent podcasting needs |
[2387.12 --> 2389.38] support more than ever he |
[2389.38 --> 2390.48] says I have two questions |
[2390.48 --> 2392.32] but and one more than more |
[2392.32 --> 2394.10] than the other all right so |
[2394.10 --> 2394.58] let's start with the |
[2394.58 --> 2396.52] important one he says in |
[2396.52 --> 2397.64] your failing at scale |
[2397.64 --> 2398.60] episode you brought up |
[2398.60 --> 2399.46] home assistant and |
[2399.46 --> 2400.64] breaking changes with z |
[2400.64 --> 2403.08] wave z wave j s etc you |
[2403.08 --> 2404.16] also mentioned waiting for |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.