Merge pull request #459 from HubSpot/logfetch_history_search

Logfetch v12 (WIP)

scripts/logfetch/entrypoint.py

@@ -3,6 +3,7 @@
 import sys
 import os
 import pkg_resources
+from datetime import datetime
 from termcolor import colored
 from fake_section_head import FakeSectionHead
 from live_logs import download_live_logs
@@ -56,6 +57,22 @@ def check_dest(args):
   if not os.path.exists(args.dest):
     os.makedirs(args.dest)
 
+def check_args(args):
+  if args.deployId and not args.requestId:
+    exit("Must specify request-id (-r) when specifying deploy-id")
+  elif not args.requestId and not args.deployId and not args.taskId:
+    exit('Must specify one of\n -t task-id\n -r request-id and -d deploy-id\n -r request-id')
+
+def convert_to_days(argument):
+    try:
+        val = int(argument)
+    except:
+      try:
+        val = (datetime.now() - datetime.strptime(argument, "%m-%d-%Y")).days
+      except:
+        exit('Start/End days value must be either a number of days or a date in format "mm-dd-yyyy"')
+    return val
+
 def fetch():
   conf_parser = argparse.ArgumentParser(version=VERSION, description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter, add_help=False)
   conf_parser.add_argument("-f", "--conf-folder", dest='conf_folder', help="specify a folder for config files to live")
@@ -70,7 +87,8 @@ def fetch():
     "chunk_size" : DEFAULT_CHUNK_SIZE,
     "dest" : DEFAULT_DEST,
     "task_count" : DEFAULT_TASK_COUNT,
-    "start_days" : DEFAULT_DAYS
+    "start_days" : DEFAULT_DAYS,
+    "end_days" : 0 #today
   }
 
   try:
@@ -91,17 +109,16 @@ def fetch():
   parser.add_argument("-n", "--num-parallel-fetches", dest="num_parallel_fetches", help="Number of fetches to make at once", type=int)
   parser.add_argument("-cs", "--chunk-size", dest="chunk_size", help="Chunk size for writing from response to filesystem", type=int)
   parser.add_argument("-u", "--singularity-uri-base", dest="singularity_uri_base", help="The base for singularity (eg. http://localhost:8080/singularity/v1)")
-  parser.add_argument("-s", "--start-days", dest="start_days", help="Search for logs no older than this many days", type=int)
-  parser.add_argument("-e", "--end-days", dest="end_days", help="Search for logs no new than this many days (defaults to None/today)", type=int)
+  parser.add_argument("-s", "--start-days", dest="start_days", help="Search for logs no older than this, can be an integer number of days or date in format 'mm-dd-yyyy'")
+  parser.add_argument("-e", "--end-days", dest="end_days", help="Search for logs no newer than this, can be an integer number of days or date in format 'mm-dd-yyyy' (defaults to None/today)")
   parser.add_argument("-l", "--log-type", dest="logtype", help="Logfile type to downlaod (ie 'access.log'), can be a glob (ie *.log)")
   parser.add_argument("-g", "--grep", dest="grep", help="Regex to grep for (normal grep syntax) or a full grep command")
 
   args = parser.parse_args(remaining_argv)
 
-  if args.deployId and not args.requestId:
-    exit("Must specify request-id (-r) when specifying deploy-id")
-  elif not args.requestId and not args.deployId and not args.taskId:
-    exit('Must specify one of\n -t task-id\n -r request-id and -d deploy-id\n -r request-id')
+  check_args(args)
+  args.start_days = convert_to_days(args.start_days)
+  args.end_days = convert_to_days(args.end_days)
 
   args.dest = os.path.expanduser(args.dest)
 
@@ -121,7 +138,8 @@ def cat():
     "chunk_size" : DEFAULT_CHUNK_SIZE,
     "dest" : DEFAULT_DEST,
     "task_count" : DEFAULT_TASK_COUNT,
-    "start_days" : DEFAULT_DAYS
+    "start_days" : DEFAULT_DAYS,
+    "end_days" : 0 #today
   }
 
   try:
@@ -142,16 +160,15 @@ def cat():
   parser.add_argument("-n", "--num-parallel-fetches", dest="num_parallel_fetches", help="Number of fetches to make at once", type=int)
   parser.add_argument("-cs", "--chunk-size", dest="chunk_size", help="Chunk size for writing from response to filesystem", type=int)
   parser.add_argument("-u", "--singularity-uri-base", dest="singularity_uri_base", help="The base for singularity (eg. http://localhost:8080/singularity/v1)")
-  parser.add_argument("-s", "--start-days", dest="start_days", help="Search for logs no older than this many days", type=int)
-  parser.add_argument("-e", "--end-days", dest="end_days", help="Search for logs no new than this many days (defaults to None/today)", type=int)
+  parser.add_argument("-s", "--start-days", dest="start_days", help="Search for logs no older than this, can be an integer number of days or date in format 'mm-dd-yyyy'")
+  parser.add_argument("-e", "--end-days", dest="end_days", help="Search for logs no newer than this, can be an integer number of days or date in format 'mm-dd-yyyy' (defaults to None/today)")
   parser.add_argument("-l", "--logtype", dest="logtype", help="Logfile type to downlaod (ie 'access.log'), can be a glob (ie *.log)")
 
   args = parser.parse_args(remaining_argv)
 
-  if args.deployId and not args.requestId:
-    exit("Must specify requestId (-r) when specifying deploy-id")
-  elif not args.requestId and not args.deployId and not args.taskId:
-    exit('Must specify one of\n -t task-id\n -r request-id and -d deploy-id\n -r request-id')
+  check_args(args)
+  args.start_days = convert_to_days(args.start_days)
+  args.end_days = convert_to_days(args.end_days)
 
   args.dest = os.path.expanduser(args.dest)
 
@@ -188,12 +205,9 @@ def tail():
 
   args = parser.parse_args(remaining_argv)
 
-  if args.deployId and not args.requestId:
-    exit("Must specify request-id (-r) when specifying deploy-id")
-  elif not args.requestId and not args.deployId and not args.taskId:
-    exit('Must specify one of\n -t task-id\n -r request-id and -d deploy-id\n -r request-id')
-  elif not args.logfile:
+  if not args.logfile:
     exit("Must specify logfile to tail (-l)")
+  check_args(args)
 
   args.dest = os.path.expanduser(args.dest)
 

scripts/logfetch/s3_logs.py

@@ -9,6 +9,7 @@
 
 TASK_FORMAT = '/task/{0}'
 S3LOGS_URI_FORMAT = '{0}/logs{1}'
+REQUEST_FORMAT = '/request/{0}'
 
 def download_s3_logs(args):
   sys.stderr.write(colored('Checking for S3 log files', 'cyan') + '\n')
@@ -37,19 +38,26 @@ def already_downloaded(dest, filename):
 
 def logs_for_all_requests(args):
   if args.taskId:
-    return get_json_response(singularity_s3logs_uri(args, args.taskId))
+    return get_json_response(s3_task_logs_uri(args, args.taskId))
   else:
     tasks = logfetch_base.tasks_for_requests(args)
     logs = []
     for task in tasks:
-      s3_logs = get_json_response(singularity_s3logs_uri(args, task))
+      s3_logs = get_json_response(s3_task_logs_uri(args, task))
       logs = logs + s3_logs if s3_logs else logs
-    return logs
+    sys.stderr.write(colored('Also searching s3 history...\n', 'magenta'))
+    for request in logfetch_base.all_requests(args):
+      s3_logs = get_json_response(s3_request_logs_uri(args, request))
+      logs = logs + s3_logs if s3_logs else logs
+    return [dict(t) for t in set([tuple(l.items()) for l in logs])]
 
 def time_from_filename(filename):
   time_string = re.search('(\d{13})', filename).group(1)
   return int(time_string[0:-3])
 
-def singularity_s3logs_uri(args, idString):
+def s3_task_logs_uri(args, idString):
   return S3LOGS_URI_FORMAT.format(logfetch_base.base_uri(args), TASK_FORMAT.format(idString))
 
+def s3_request_logs_uri(args, idString):
+  return S3LOGS_URI_FORMAT.format(logfetch_base.base_uri(args), REQUEST_FORMAT.format(idString))
+

scripts/setup.py

@@ -10,7 +10,7 @@
 
 setup(
     name='singularity-logfetch',
-    version='0.11.0',
+    version='0.12.0',
     description='Singularity log fetching and searching',
     author="HubSpot",
     author_email='[email protected]',