oar-team / batsim

Batsim: Infrastructure simulator for job and I/O scheduling
GNU Lesser General Public License v3.0
30 stars 15 forks source link

Invalid _pstate_changes.csv files #25

Closed mpoquet closed 3 years ago

mpoquet commented 7 years ago

While running some experiments on Grid'5000, I got two invalid _pstate_changes.csv files (out of 1282).

Data: buggy_pstate_changes.zip

Some IO problem probably occurred during the execution of these instances. But it might be possible to improve the exportation system such that the risk to see this kind of problem is reduced.

Currently running these instances again to make sure the problem is not more grave. Update: got the same problem after a new execution of the instances, the problem seems deterministic regarding Batsim inputs!

mpoquet commented 7 years ago

Detected those files with the following python command:

filenames = glob.glob('./*_pstate_changes.csv')

for f in filenames:
    try:
        data = pd.read_csv(f)
    except:
        print('cannot read {}'.format(f))
mpoquet commented 7 years ago

The following script displays (maybe not every) invalid lines.

import re
import glob

filenames = glob.glob('./*_pstate_changes.csv')
r = re.compile('''^[^,]+,[^,]+,[^,]+$''')

for fname in filenames:
    with open(fname) as f:
        line_nb = 1
        for line in f:
            line = line.strip()
            if r.match(line) is None:
                print('{f}:{l}... "{c}"'.format(f=fname, l=line_nb, c=line))
            line_nb += 1

Result:

./90e1b789_pstate_changes.csv:14993... "4.53444e+06,117,4.54.53445e+06,117,13"
./90e1b789_pstate_changes.csv:18539... "5.89263e+06,7-105.85.89272e+06,4,-2"
./90e1b789_pstate_changes.csv:126461... "2 32,-1"
./224b32a4_pstate_changes.csv:17309... "1.82085e+06,34-35 38-65,1.81.82086e+06,34-35 38-65,13"
./224b32a4_pstate_changes.csv:20182... "2.36649e+06,43-46 120-122.3662.36649e+06,43-46 120-121,13"
./224b32a4_pstate_changes.csv:26022... "3.23396e+06,44-47 91-1023.233.2355e+06,0-3,-1"
./224b32a4_pstate_changes.csv:33923... "4.2114e+06,0-3 71-79 88-4.2114e4.21141e+06,4 10 31-37 48-70,13"
./224b32a4_pstate_changes.csv:36489... "4.584.58614e+06,41-4.58614e+06,122,13"
./224b32a4_pstate_changes.csv:41350... "5.02998e+06,2-17 34-47 75.02998e5.02998e+06,18-33,-1"
./224b32a4_pstate_changes.csv:48524... "5.63634e+06,97 100-101 15.63635.63635e+06,14-15 38-39 72-73 80-84 92-96,13"
./224b32a4_pstate_changes.csv:190715... "2.39788e+07,53-2.39774e+072.39788e+07,0-9 42-44,0"
./224b32a4_pstate_changes.csv:195702... "2.44957e2.44928e+07,2.44957e+07,0-31,-1"
./224b32a4_pstate_changes.csv:198302... "2.49832.49675e+07,712.49833e+07,68-70 103-104,0"
./224b32a4_pstate_changes.csv:200944... "2.5284e+07,13-29 33 42.52864e+07,13-29 33 46-47,0"
./224b32a4_pstate_changes.csv:203362... "2.55185e+07,102.55016e+07,107,13"
./224b32a4_pstate_changes.csv:206164... "2.58601e+07,7 11 33-34 2.58569e+07,7 35-37,13"
./224b32a4_pstate_changes.csv:206165... "2.58574e+07,9,-2.58603e+07,7 11 33-34 37-40,0"
./224b32a4_pstate_changes.csv:208721... "2.60793e+07,412.6083e+07,41-44 49 61-68 70-71 73-76 79-82 86 89-104,0"
./224b32a4_pstate_changes.csv:210932... "2.63797e+07,94,2.63729e+07,25-26 45 47-48 119,13"
./224b32a4_pstate_changes.csv:210933... "2.63731e+07,46 92-93 992.63797e+07,88,-2"
./224b32a4_pstate_changes.csv:213570... "2.65234e+07,16-2.65217e+07,0-15,-1"
./224b32a4_pstate_changes.csv:213572... "2.65219e+07,2.65235e+07,16-23 113 116-122,0"
./224b32a4_pstate_changes.csv:221586... "2.76408e+07,84 93 109-114,-22.76268e+07,117-118,-2"
./224b32a4_pstate_changes.csv:224513... "2.80387e+07,51 66,12.80601e+07,0 13 51,13"
./224b32a4_pstate_changes.csv:227456... "2.83643e+07,49-50 62 62.83628e+07,63-64 68-69 93-95 98-103 106 113-114,0"
./224b32a4_pstate_changes.csv:230407... "2.87928e+07,0-2 52.87815e+07,7,-1"
./224b32a4_pstate_changes.csv:233104... "2.94078e+07,2.94499e+07,111-114,-1"
./224b32a4_pstate_changes.csv:235454... "2.95795e+07,12.9582e+07,16-19,-1"
./224b32a4_pstate_changes.csv:238048... "2.9834e+07,8-21 56-2.9831e+07,8-14,0"
./224b32a4_pstate_changes.csv:238050... "2.9831e+07,12.98341e+07,8-39 56-59,-2"
./224b32a4_pstate_changes.csv:240644... "3.00814e+07,0-14 33-36 99-101 118-123.00799e+07,0-15 32-33 36-39 82-86 94-96,-2"
./224b32a4_pstate_changes.csv:240645... "3.00799e+07,34-3.00816e+07,0-14,0"
./224b32a4_pstate_changes.csv:243040... "3.01775e+07,82 92-110 126,-23.01744e+07,0-15,-2"
./224b32a4_pstate_changes.csv:243042... "3.01744e+07,93.01775e+07,1-15 82 92-99,-1"
./224b32a4_pstate_changes.csv:245727... "3.05009e+07,96-103.05175e+07,100-103 105,-2"
./224b32a4_pstate_changes.csv:248060... "3.07052e+07,104-107,133.07035e+07,124-125,13"
./224b32a4_pstate_changes.csv:250480... "3.08777e+07,17-28 3.08761e+07,47-54,-2"
./224b32a4_pstate_changes.csv:250482... "3.08761e+07,47-53.08777e+07,120-121,13"
./224b32a4_pstate_changes.csv:252799... "3.10208e+07,0,-3.10101e+07,19-24 31-32 34-43 98-103,0"
./224b32a4_pstate_changes.csv:252801... "3.10101e+07,903.1021e+07,0,0"
./224b32a4_pstate_changes.csv:257223... "3.13478e+07,42-43.13577e+07,112-113,-1"
./224b32a4_pstate_changes.csv:259966... "3.16462e+07,0 4-5 12-13 16-18 24-39 41-46 56-63 3.1701e+07,79-110,0"
./224b32a4_pstate_changes.csv:262527... "3.20923e+07,64-63.20842e+07,21-27 32-34 52 56-57 59-68,-2"
./224b32a4_pstate_changes.csv:265219... "3.25775e+07,48-793.25816e+07,48-79,-2"
./224b32a4_pstate_changes.csv:268005... "3.31198e+07,78 83.31161e+07,2-3,0"
./224b32a4_pstate_changes.csv:272396... "3.34909e+07,0-3 53.34883e+07,0-3 5-6 8-9,-2"
./224b32a4_pstate_changes.csv:274950... "3.36368e+07,8-11,3.36397e+07,8-11,13"
./224b32a4_pstate_changes.csv:277521... "3.38642e+07,93.38502e+07,38-53 62-77,-1"
./224b32a4_pstate_changes.csv:285560... "3.48365e+07,2-3,13.48467e+07,91,0"
./224b32a4_pstate_changes.csv:288268... "3.53387e+07,3.53108e+07,13-44,13"
./224b32a4_pstate_changes.csv:290908... "3.55301e+07,0-3.55332e+07,0-7,-2"
./224b32a4_pstate_changes.csv:293831... "3.58305e+07,8-3.58266e+07,0-3,-1"
./224b32a4_pstate_changes.csv:299028... "3.62843e+07,83-863.62942e+07,87-88 90 98-100,0"
./224b32a4_pstate_changes.csv:301526... "3.65626e+07,60-63 77 83-88 90-92 96 3.65949e+07,52-55,0"
./224b32a4_pstate_changes.csv:307330... "3.77216e+07,77-92 102-1163.77259e+07,2-3 77-92 102-115,-1"
./224b32a4_pstate_changes.csv:310402... "3.81732e+07,64-73 79-80 82-893.81589e+07,64-68,-2"
./224b32a4_pstate_changes.csv:319200... "3.98334e+07,3.98399e+07,32-33,0"
./224b32a4_pstate_changes.csv:324581... "4.05427e+07,0-3 18-24 29-35,-24.05386e+07,0 13-15 17-24,0"
./224b32a4_pstate_changes.csv:327529... "4.06373e+07,0-5,04.06342e+07,0-7 18-24 29-31,-2"
./224b32a4_pstate_changes.csv:330031... "4.09131e+07,18-19 26-31 48-49 75-76 120-1234.09113e+07,120-123,13"
./224b32a4_pstate_changes.csv:332663... "4.10853e+07,22-23 30-31,4.10909e+07,24 49,0"
./224b32a4_pstate_changes.csv:334809... "4.12999e+07,95-96 103-104 1184.12895e+07,22-25 30-57,-2"
./224b32a4_pstate_changes.csv:334813... "4.12897e+07,24.13001e+07,22-25 27-44 79-81 95-96 103-104 118-120,0"
./224b32a4_pstate_changes.csv:339521... "4.16817e+07,4.16791e+07,2-3 15-25 27-29,0"
./224b32a4_pstate_changes.csv:339524... "4.16792e+07,92-99 14.16819e+07,2-3 15-16 23-25 27-29 88-93,0"
./224b32a4_pstate_changes.csv:344458... "4.20136e+07,54-55 4.20109e+07,48-49,0"
./224b32a4_pstate_changes.csv:346824... "4.22999e+07,49-50 92-98 14.22782e+07,0-11 18-26 32-38 43-56 122-123,-2"
./224b32a4_pstate_changes.csv:346827... "4.22782e+07,0-11 18-26 32-38,4.23e+07,51-56 108-109 119-122,0"
./224b32a4_pstate_changes.csv:349222... "4.25673e+07,32-41 48-49 96-974.25648e+07,42-47 96-97,-1"
./224b32a4_pstate_changes.csv:351849... "4.30257e+07,0-31,-24.29989e+07,111-112,-2"
./224b32a4_pstate_changes.csv:354641... "4.34185e+07,12-4.34128e+07,10-11,-2"
./224b32a4_pstate_changes.csv:357067... "4.40696e+07,100-101 105 4.40867e+07,93-95,13"
./224b32a4_pstate_changes.csv:365331... "4.52885e+07,32-44.52924e+07,40 42-52 121-124,0"
./224b32a4_pstate_changes.csv:369807... "4.57109e+07,40-41 53-4.57088e+07,86-87 103 112 122,13"
./224b32a4_pstate_changes.csv:369809... "4.57088e+07,40-41 53-57 72-75 86-87 103 112 1224.57111e+07,86-87 103 112 122,-2"
./224b32a4_pstate_changes.csv:377795... "4.66254e+07,4.66294e+07,79-85 96-120,13"
./224b32a4_pstate_changes.csv:380469... "4.72012e+07,1124.7189e+07,100-107,13"
./224b32a4_pstate_changes.csv:380473... "4.71895e+07,974.72014e+07,113-114,-2"
./224b32a4_pstate_changes.csv:388010... "4.82126e+07,44 49-50 54.82094e+07,40-43 45-48 51-58,0"
./224b32a4_pstate_changes.csv:388012... "4.82094e+07,44 494.82128e+07,44 49-50 59-60 65-75,-2"
./224b32a4_pstate_changes.csv:390787... "4.87429e+07,4.86961e+07,98,-1"
./224b32a4_pstate_changes.csv:393638... "4.95328e+07,74.95306e+07,109-115 127,13"
./224b32a4_pstate_changes.csv:393640... "4.95307e+07,40-59 72 76-77 814.9533e+07,109-115 127,13"
./224b32a4_pstate_changes.csv:396253... "4.98141e+07,4.98112e+07,0-15 32-39 72-79,0"
./224b32a4_pstate_changes.csv:396257... "4.98112e+07,0-14.98143e+07,0-15 32-39 72-79,0"
./224b32a4_pstate_changes.csv:399291... "5.09969e+07,84-5.09607e+07,0-31,-1"
./224b32a4_pstate_changes.csv:402217... "5.17108e+07,0-1 5.17163e+07,2-5,-1"
./224b32a4_pstate_changes.csv:410884... "5.38004e+07,11 15.3797e+07,6-9,-2"
./224b32a4_pstate_changes.csv:416609... "5.4967e+07,7-9 86 88-89 106-105.49698e+07,7 10-11 72-76 85 110-111,0"
./224b32a4_pstate_changes.csv:418999... "5.51439e+07,57-58 61-66,5.51472e+07,57-58 61-66,-2"
./224b32a4_pstate_changes.csv:421664... "5.56271e+07,111-15.56242e+07,122-127,13"
./224b32a4_pstate_changes.csv:421666... "5.56243e+07,43-47 95.56271e+07,39-42,13"
./224b32a4_pstate_changes.csv:428603... "5.62656e+07,4-7 12-19 64-5.62638e+07,73-79 93-95 102 111-115,0"
./224b32a4_pstate_changes.csv:430759... "5.66959e+07,5.67044e+07,40-42,-1"
./224b32a4_pstate_changes.csv:432768... "5.69795e+07,92 96-99 111-113 115-117,135.69779e+07,23 64 66-69 92 96-99 111-1135.69795e+07,23 64 66-69 92 96-99 111-113 115-120,-1"
./224b32a4_pstate_changes.csv:434333... "5.70266e+07,20-22 64-5.70248e+07,23-24 30-31 62-63,-25.70268e+07,64-66 92 96-99 111-113 115-120,-2"
./224b32a4_pstate_changes.csv:436972... "5.73169e+07,22-5.73403e+07,20,-1"
./224b32a4_pstate_changes.csv:441556... "5.77766e+07,25 30 32-34 46-48 575.77741e+05.77766e+07,25 30 32-34 46-48,-1"
mpoquet commented 7 years ago

Some data to reproduce the error (not clean yet). reproduce.zip

mickours commented 6 years ago

It looks like 2 IO streams are mixed. Maybe stream the outputs correctly to the file at runtime instead of at the end of the simulation would fix the issue?

mpoquet commented 6 years ago

The issue is old, the exportation of data has been improved since the issue has been opened.

We should check if the problem still exists...

mpoquet commented 5 years ago

We should really check this with current Batsim.

Mommessc commented 4 years ago

We definitely should check this.

mpoquet commented 3 years ago

What about checking this @Mommessc? :D

Mommessc commented 3 years ago

After struggling with your very old data I managed to run the batsim simulation but the python script did not report any faulty line.

mpoquet commented 3 years ago

OK, I close this issue, this has probably been fixed a long time ago.