Our collectors run long at regular intervals; in particular every two hours, and to lesser extents every hour and half hour. Here's a graph showing how long each collection cycle lasts on one of the collection machines: <a href="http://i.imgur.com/xaZJ5.png">http://i.imgur.com/xaZJ5.png</a> – note the regular spikes.<br>
<meta charset="utf-8"><br><div>Most RRD's consolidate every 30 minutes, 2 hours, and 24 hours; see the bottom for a sample `rrd info`. Our current theory is that the RRD consolidations are causing these long runtimes. If that's the case, is there a way to evenly stagger the consolidations over time so we can better distribute RRD update load?</div>
<div><div><br></div><div>Thanks,</div><div>Joshua</div><div><br></div><div><div>filename = "/rrd/router/<a href="http://cr01.ptleorte.integra.net/tengigabitethernet134.rrd">cr01.ptleorte.integra.net/tengigabitethernet134.rrd</a>" </div>
<div>rrd_version = "0003" </div><div>step = 300 </div>
<div>last_update = 1304228713 </div><div>ds[ds0].type = "COUNTER" </div>
<div>ds[ds0].minimal_heartbeat = 600 </div><div>ds[ds0].min = 0.0000000000e+00 </div>
<div>ds[ds0].max = 1.2500000000e+09 </div><div>ds[ds0].last_ds = "1596044569532963" </div>
<div>ds[ds0].value = 4.0248335433e+08 </div><div>ds[ds0].unknown_sec = 0 </div>
<div>ds[ds1].type = "COUNTER" </div><div>ds[ds1].minimal_heartbeat = 600 </div>
<div>ds[ds1].min = 0.0000000000e+00 </div><div>ds[ds1].max = 1.2500000000e+09 </div>
<div>ds[ds1].last_ds = "3460406816844600" </div><div>ds[ds1].value = 8.9596753966e+08 </div>
<div>ds[ds1].unknown_sec = 0 </div><div>rra[0].cf = "AVERAGE" </div>
<div>rra[0].rows = 600 </div><div>rra[0].pdp_per_row = 1 </div>
<div>rra[0].xff = 5.0000000000e-01 </div><div>rra[0].cdp_prep[0].value = NaN </div>
<div>rra[0].cdp_prep[0].unknown_datapoints = 0 </div><div>rra[0].cdp_prep[1].value = NaN </div>
<div>rra[0].cdp_prep[1].unknown_datapoints = 0 </div><div>rra[1].cf = "AVERAGE" </div>
<div>rra[1].rows = 600 </div><div>rra[1].pdp_per_row = 6 </div>
<div>rra[1].xff = 5.0000000000e-01 </div><div>rra[1].cdp_prep[0].value = 9.4104250250e+07 </div>
<div>rra[1].cdp_prep[0].unknown_datapoints = 0 </div><div>rra[1].cdp_prep[1].value = 2.0174889583e+08 </div>
<div>rra[1].cdp_prep[1].unknown_datapoints = 0 </div><div>rra[2].cf = "AVERAGE" </div>
<div>rra[2].rows = 600 </div><div>rra[2].pdp_per_row = 24 </div>
<div>rra[2].xff = 5.0000000000e-01 </div><div>rra[2].cdp_prep[0].value = 6.5449761744e+08 </div>
<div>rra[2].cdp_prep[0].unknown_datapoints = 0 </div><div>rra[2].cdp_prep[1].value = 1.4734297081e+09 </div>
<div>rra[2].cdp_prep[1].unknown_datapoints = 0 </div><div>rra[3].cf = "AVERAGE" </div>
<div>rra[3].rows = 732 </div><div>rra[3].pdp_per_row = 288 </div>
<div>rra[3].xff = 5.0000000000e-01 </div><div>rra[3].cdp_prep[0].value = 2.2692529674e+09 </div>
<div>rra[3].cdp_prep[0].unknown_datapoints = 3 </div><div>rra[3].cdp_prep[1].value = 4.7002069004e+09 </div>
<div>rra[3].cdp_prep[1].unknown_datapoints = 3 </div><div>rra[4].cf = "MAX" </div>
<div>rra[4].rows = 600 </div><div>rra[4].pdp_per_row = 1 </div>
<div>rra[4].xff = 5.0000000000e-01 </div><div>rra[4].cdp_prep[0].value = NaN </div>
<div>rra[4].cdp_prep[0].unknown_datapoints = 0 </div><div>rra[4].cdp_prep[1].value = NaN </div>
<div>rra[4].cdp_prep[1].unknown_datapoints = 0 </div><div>rra[5].cf = "MAX" </div>
<div>rra[5].rows = 600 </div><div>rra[5].pdp_per_row = 6 </div>
<div>rra[5].xff = 5.0000000000e-01 </div><div>rra[5].cdp_prep[0].value = 3.2405792329e+07 </div>
<div>rra[5].cdp_prep[0].unknown_datapoints = 0 </div><div>rra[5].cdp_prep[1].value = 6.9813629778e+07 </div>
<div>rra[5].cdp_prep[1].unknown_datapoints = 0 </div><div>rra[6].cf = "MAX" </div>
<div>rra[6].rows = 600 </div><div>rra[6].pdp_per_row = 24 </div>
<div>rra[6].xff = 5.0000000000e-01 </div><div>rra[6].cdp_prep[0].value = 3.4089842030e+07 </div>
<div>rra[6].cdp_prep[0].unknown_datapoints = 0 </div><div>rra[6].cdp_prep[1].value = 7.6745619740e+07 </div>
<div>rra[6].cdp_prep[1].unknown_datapoints = 0 </div><div>rra[7].cf = "MAX" </div>
<div>rra[7].rows = 732 </div><div>rra[7].pdp_per_row = 288 </div>
<div>rra[7].xff = 5.0000000000e-01 </div><div>rra[7].cdp_prep[0].value = 4.4271024386e+07 </div>
<div>rra[7].cdp_prep[0].unknown_datapoints = 3 </div><div>rra[7].cdp_prep[1].value = 8.8648080465e+07 </div>
<div>rra[7].cdp_prep[1].unknown_datapoints = 3 </div></div><div><br></div></div>