Not Done, But Done For Now

I spent some more time on it, and I figured something out.

I looked at the data, and instead of getting 1 2 3 4 NULL NULL 5 6 7, I was getting 1 2 3 4 NULL NULL 7 1 2, starting at the beginning again. So, I figured out how to do loops and made a series of vectors, containing the dates in one, and load averages per VM.

	#!/usr/bin/env Rscript
	require( Cairo , quietly=TRUE )
	# read in file
	logfile = read.table('.uptime.log')
	# set column names
	colnames(logfile)=c('date','time','log','host','colon','load','x','y')
	# create new table with just the columns we need
	needed = logfile[c("host","date","time","load")]
	# break into four tables: genomics,genomics-apps, genomics-test, genomics-db
	needed$datetime <- paste( as.character(needed$date) , as.character(needed$time) )
	# remove unnecessary seconds from time
	needed$datetime <- sub('......$','',needed$datetime)
	needed$datetime <- sub('/','',needed$datetime)
	needed$datetime <- sub('/','',needed$datetime)
	needed$datetime <- sub(' ','',needed$datetime)
	needed$datetime <- sub('$','00',needed$datetime)
	# instead of 'YYYY/MM/DD hh:mm:ss', we have 'YYYYMMDDhh00'
	# remove unnecessary columns
	final <- needed[c("host","datetime","load")]
	# all dates in data table
	dates <- unique(final$datetime)
	# all tables handled the same:
	# subset gives table of those with correct hostname
	# then cut host column
	genomics <- subset(final, host=='genomics')
	genomics <- genomics[c("datetime","load")]
	genomicsapps <- subset(final, host=='genomics-apps')
	genomicsapps <- genomicsapps[c("datetime","load")]
	genomicsdb <- subset(final, host=='genomics-db')
	genomicsdb <- genomicsdb[c("datetime","load")]
	genomicstest <- subset(final, host=='genomics-test')
	genomicstest <- genomicstest[c("datetime","load")]
	# I create an empty vector
	# for each date in dates, which contains all datetimes
	# I get v, which is either the value or NA
	# if it is NA, which I cannot deal with,
	# I make it -1, which I can
	# and push it onto the vector
	genomics.v <- vector()
	for( k in dates ) {
	v <- genomics$load[ match( k,genomics$datetime ) ]
	if ( is.na(v) ) { v <- -1 }
	genomics.v <- c(genomics.v,v)
	}
	genomicsapps.v <- vector()
	for( k in dates ) {
	v <- genomicsapps$load[ match( k,genomicsapps$datetime ) ]
	if ( is.na(v) ) { v <- -1 }
	genomicsapps.v <- c(genomicsapps.v,v)
	}
	genomicsdb.v <- vector()
	for( k in dates ) {
	v <- genomicsdb$load[ match( k,genomicsdb$datetime ) ]
	if ( is.na(v) ) { v <- -1 }
	genomicsdb.v <- c(genomicsdb.v,v)
	}
	genomicstest.v <- vector()
	for( k in dates ) {
	v <- genomicstest$load[ match( k,genomicstest$datetime ) ]
	if ( is.na(v) ) { v <- -1 }
	genomicstest.v <- c(genomicstest.v,v)
	}
	# this pulls things back to the last four days, which is all I want ATM
	fourdays = 96
	label_range=seq(0,96,8)
	dates <- tail( dates , n=fourdays )
	genomics.v <- tail( genomics.v , n=fourdays )
	genomicsapps.v <- tail( genomicsapps.v , n=fourdays )
	genomicsdb.v <- tail( genomicsdb.v , n=fourdays )
	genomicstest.v <- tail( genomicstest.v , n=fourdays )
	# names and colors for the plot
	names = c("genomics", "genomics-apps", "genomics-db", "genomics-test" )
	p_cols <- c( "blue" , "red" , "orange" , "green" )
	yrange <-range(-1, 5, genomics.v , genomicstest.v , genomicsdb.v , genomicsapps.v )
	# from here on, there's plotting administrivia
	CairoPNG( filename="load.png"
	, width = 600
	, height = 400
	, pointsize = 12
	)
	plot ( genomics.v
	, col=p_cols[1]
	, type="l"
	, xlab="time"
	, ylab="load avg"
	, ylim=yrange
	, las=1
	, xaxt="n"
	, sub="hours ago"
	)
	lines( genomicsapps.v
	, col=p_cols[2]
	, type="l"
	)
	lines( genomicsdb.v
	, col=p_cols[3]
	, type="l"
	)
	lines( genomicstest.v
	, col=p_cols[4]
	, type="l"
	)
	# where to put the legend box, based on x/y of plot
	legend_x <- 4
	legend_y <- 4.5
	legend(
	legend_x, legend_y, names, col=p_cols ,
	cex=0.8 , lty=c(1,1,1)
	)
	axis( 1 , at=label_range , labels=rev(label_range) )
	box()
	title( main="Genomics VM Load Average" )

view raw uptime.R hosted with ❤ by GitHub

Lev suggested that this is not how a real R person would do it. True. But this works, and I know how to plot vectors but not data tables. So, a few more changes (having the date in the title is good) and I can finish it up and put it into my workflow. Yay me.