Merge branch 'master' into mayday

.cspell.json

@@ -13,7 +13,6 @@
         "erpnext",
         "parenttype",
         "parentfield",
-        "subsription",
         "sbool",
         "binlog",
         "ifnull",
@@ -68,6 +67,9 @@
         "nvme",
         "nofail"
 
+        "hset",
+        "serializability",
+        "oom"
     ],
     "allowCompoundWords": true,
     "ignorePaths": [
@@ -82,4 +84,4 @@
         "requirements.txt",
         "press/utils/country_timezone.py"
     ]
-}
+}

press/api/monitoring.py

@@ -1,4 +1,3 @@
-# -*- coding: utf-8 -*-
 # Copyright (c) 2021, Frappe and contributors
 # For license information, please see license.txt
 
@@ -7,15 +6,11 @@
 
 import frappe
 
+from press.exceptions import AlertRuleNotEnabled
 from press.utils import log_error
 
 
-@frappe.whitelist(allow_guest=True)
-def targets(token):
-	monitor_token = frappe.db.get_single_value("Press Settings", "monitor_token")
-	if token != monitor_token:
-		return
-
+def get_benches():
 	self_hosted_stand_alone_servers = frappe.get_all(
 		"Server",
 		{"is_standalone": True, "is_self_hosted": True, "status": "Active"},
@@ -39,18 +34,18 @@ def targets(token):
 		)
 	}
 	benches = []
-	for bench_name, sites in groupby(sites, lambda x: x.bench):
+	for bench_name, _sites in groupby(sites, lambda x: x.bench):
 		bench = bench_map[bench_name]
-		bench.update({"sites": [site.name for site in sites]})
+		bench.update({"sites": [site.name for site in _sites]})
 		benches.append(bench)
 
+	return benches
+
+
+def get_clusters():
 	servers = {}
-	servers["proxy"] = frappe.get_all(
-		"Proxy Server", {"status": ("!=", "Archived")}, ["name", "cluster"]
-	)
-	servers["app"] = frappe.get_all(
-		"Server", {"status": ("!=", "Archived")}, ["name", "cluster"]
-	)
+	servers["proxy"] = frappe.get_all("Proxy Server", {"status": ("!=", "Archived")}, ["name", "cluster"])
+	servers["app"] = frappe.get_all("Server", {"status": ("!=", "Archived")}, ["name", "cluster"])
 	servers["database"] = frappe.get_all(
 		"Database Server", {"status": ("!=", "Archived")}, ["name", "cluster"]
 	)
@@ -69,10 +64,16 @@ def targets(token):
 					for job in job_map[server_type]:
 						cluster["jobs"].setdefault(job, []).append(server.name)
 
-	domains = frappe.get_all(
+	return clusters
+
+
+def get_domains():
+	return frappe.get_all(
 		"Site Domain", ["name", "site"], {"tls_certificate": ("is", "set")}, order_by="name"
 	)
 
+
+def get_tls():
 	tls = []
 	server_types = [
 		"Server",
@@ -87,13 +88,22 @@ def targets(token):
 	for server_type in server_types:
 		tls += frappe.get_all(server_type, {"status": ("!=", "Archived")}, ["name"])
 
-	return {"benches": benches, "clusters": clusters, "domains": domains, "tls": tls}
+	return tls
+
+
+@frappe.whitelist(allow_guest=True)
+def targets(token):
+	monitor_token = frappe.db.get_single_value("Press Settings", "monitor_token")
+	if token != monitor_token:
+		return None
+
+	return {"benches": get_benches(), "clusters": get_clusters(), "domains": get_domains(), "tls": get_tls()}
 
 
 @frappe.whitelist(allow_guest=True, xss_safe=True)
 def alert(*args, **kwargs):
 	try:
-		user = frappe.session.user
+		user = str(frappe.session.user)
 		frappe.set_user("Administrator")
 
 		doc = frappe.get_doc(
@@ -103,8 +113,10 @@ def alert(*args, **kwargs):
 			}
 		)
 		doc.insert()
+	except AlertRuleNotEnabled:
+		pass
 	except Exception:
 		log_error("Alertmanager Webhook Error", args=args, kwargs=kwargs)
-		raise Exception
+		raise
 	finally:
 		frappe.set_user(user)

press/exceptions.py

@@ -39,3 +39,7 @@ class ConflictingCAARecord(ValidationError):
 
 class TeamHeaderNotInRequestError(ValidationError):
 	pass
+
+
+class AlertRuleNotEnabled(ValidationError):
+	pass

press/press/doctype/agent_job/agent_job.py

@@ -343,6 +343,10 @@ def failed_because_of_agent_update(self) -> bool:
 			return True
 		return False
 
+	@property
+	def on_public_server(self):
+		return bool(frappe.db.get_value(self.server_type, self.server, "public"))
+
 
 def job_detail(job):
 	job = frappe.get_doc("Agent Job", job)

press/press/doctype/agent_job/agent_job_notifications.py

@@ -57,13 +57,19 @@ class JobErr(Enum):
 	ROW_SIZE_TOO_LARGE = auto()
 	DATA_TRUNCATED_FOR_COLUMN = auto()
 	BROKEN_PIPE_ERR = auto()
+	CANT_CONNECT_TO_MYSQL = auto()
+	GZIP_TAR_ERR = auto()
+	UNKNOWN_COMMAND_HYPHEN = auto()
 
 
 DOC_URLS = {
 	JobErr.OOM: "https://frappecloud.com/docs/common-issues/oom-issues",
 	JobErr.ROW_SIZE_TOO_LARGE: "https://frappecloud.com/docs/faq/site#row-size-too-large-error-on-migrate",
 	JobErr.DATA_TRUNCATED_FOR_COLUMN: "https://frappecloud.com/docs/faq/site#data-truncated-for-column",
 	JobErr.BROKEN_PIPE_ERR: None,
+	JobErr.CANT_CONNECT_TO_MYSQL: "https://frappecloud.com/docs/cant-connect-to-mysql-server",
+	JobErr.GZIP_TAR_ERR: "https://frappecloud.com/docs/sites/migrate-an-existing-site#tar-gzip-command-fails-with-unexpected-eof",
+	JobErr.UNKNOWN_COMMAND_HYPHEN: "https://frappecloud.com/docs/unknown-command-",
 }
 
 
@@ -88,11 +94,15 @@ def handlers() -> list[UserAddressableHandlerTuple]:
 	Due to this order of the tuples matter.
 	"""
 	return [
-		("returned non-zero exit status 137", update_with_oom_error),
-		("returned non-zero exit status 143", update_with_oom_error),
-		("Row size too large", update_with_row_size_too_large_error),
-		("Data truncated for column", update_with_data_truncated_for_column_error),
+		("returned non-zero exit status 137", update_with_oom_err),
+		("returned non-zero exit status 143", update_with_oom_err),
+		("Row size too large", update_with_row_size_too_large_err),
+		("Data truncated for column", update_with_data_truncated_for_column_err),
 		("BrokenPipeError", update_with_broken_pipe_err),
+		("ERROR 2002 (HY000)", update_with_cant_connect_to_mysql_err),
+		("gzip: stdin: unexpected end of file", update_with_gzip_tar_err),
+		("tar: Unexpected EOF in archive", update_with_gzip_tar_err),
+		("Unknown command '\\-'.", update_with_unknown_command_hyphen_err),
 	]
 
 
@@ -165,7 +175,7 @@ def get_details(job: AgentJob, title: str, message: str) -> Details:
 	return details
 
 
-def update_with_oom_error(
+def update_with_oom_err(
 	details: Details,
 	job: AgentJob,
 ):
@@ -192,7 +202,7 @@ def update_with_oom_error(
 	return False
 
 
-def update_with_row_size_too_large_error(details: Details, job: AgentJob):
+def update_with_row_size_too_large_err(details: Details, job: AgentJob):
 	details["title"] = "Row size too large error"
 
 	details[
@@ -207,7 +217,7 @@ def update_with_row_size_too_large_error(details: Details, job: AgentJob):
 	return True
 
 
-def update_with_data_truncated_for_column_error(details: Details, job: AgentJob):
+def update_with_data_truncated_for_column_err(details: Details, job: AgentJob):
 	details["title"] = "Data truncated for column error"
 
 	details[
@@ -237,6 +247,49 @@ def update_with_broken_pipe_err(details: Details, job: AgentJob):
 	return True
 
 
+def update_with_cant_connect_to_mysql_err(details: Details, job: AgentJob):
+	details["title"] = "Can't connect to MySQL server"
+
+	suggestion = "To rectify this issue, please follow the steps mentioned in <i>Help</i>."
+	if job.on_public_server:
+		suggestion = "Please raise a support ticket if the issue persists."
+
+	details[
+		"message"
+	] = f"""<p>The server couldn't connect to MySQL server during the job. This likely happened as the mysql server restarted as it didn't have sufficient memory for the operation</p>
+	<p>{suggestion}</p>
+	"""
+
+	details["assistance_url"] = DOC_URLS[JobErr.CANT_CONNECT_TO_MYSQL]
+
+	return True
+
+
+def update_with_gzip_tar_err(details: Details, job: AgentJob):
+	details["title"] = "Corrupt backup file"
+
+	details["message"] = f"""<p>An error occurred when extracting the backup to {job.site}.</p>
+	<p>To rectify this issue, please follow the steps mentioned in <i>Help</i>.</p>
+	"""
+
+	details["assistance_url"] = DOC_URLS[JobErr.GZIP_TAR_ERR]
+
+	return True
+
+
+def update_with_unknown_command_hyphen_err(details: Details, job: AgentJob):
+	details["title"] = "Incompatible site backup"
+
+	details["message"] = f"""<p>An error occurred when extracting the backup to {job.site}.</p>
+	<p>This happens when the backup is taken from a later version of MariaDB and restored on a older version.</p>
+	<p>To rectify this issue, please follow the steps mentioned in <i>Help</i>.</p>
+	"""
+
+	details["assistance_url"] = DOC_URLS[JobErr.UNKNOWN_COMMAND_HYPHEN]
+
+	return True
+
+
 def get_default_title(job: AgentJob) -> str:
 	if job.job_type == "Update Site Migrate":
 		return "Site Migrate"

press/press/doctype/alertmanager_webhook_log/alertmanager_webhook_log.py

@@ -12,6 +12,7 @@
 from frappe.utils.background_jobs import enqueue_doc
 from frappe.utils.data import add_to_date
 
+from press.exceptions import AlertRuleNotEnabled
 from press.press.doctype.incident.incident import INCIDENT_ALERT, INCIDENT_SCOPE
 from press.press.doctype.telegram_message.telegram_message import TelegramMessage
 from press.utils import log_error
@@ -78,7 +79,9 @@ def clear_old_logs(days=10):
 
 	def validate(self):
 		self.parsed = json.loads(self.payload)
-		self.alert = self.parsed["groupLabels"]["alertname"]
+		self.alert = self.parsed["groupLabels"].get("alertname")
+		if not self.alert:
+			raise AlertRuleNotEnabled("No alertname found in groupLabels")
 		self.status = self.parsed["status"].capitalize()
 		self.severity = self.parsed["commonLabels"]["severity"].capitalize()
 		self.group_key = self.parsed["groupKey"]

press/press/doctype/site/site.py

@@ -594,8 +594,8 @@ def rename(self, new_name: str):
 			# update the subscription config while renaming the standby site
 			self.update_config_preview()
 			site_config = json.loads(self.config)
-			subsription_config = site_config.get("subscription", {})
-			job = agent.rename_site(self, new_name, create_user, config={"subscription": subsription_config})
+			subscription_config = site_config.get("subscription", {})
+			job = agent.rename_site(self, new_name, create_user, config={"subscription": subscription_config})
 			self.flags.rename_site_agent_job_name = job.name
 		else:
 			agent.rename_site(self, new_name)
@@ -2320,7 +2320,6 @@ def get_database_performance_report(self):
 		from press.press.report.mariadb_slow_queries.mariadb_slow_queries import get_data as get_slow_queries
 
 		agent = Agent(self.server)
-		result = agent.get_summarized_performance_report_of_database(self)
 		# fetch slow queries of last 7 days
 		slow_queries = get_slow_queries(
 			frappe._dict(
@@ -2334,27 +2333,40 @@ def get_database_performance_report(self):
 				}
 			)
 		)
-		# remove `parent` & `creation` indexes from unused_indexes
-		result["unused_indexes"] = [
-			index
-			for index in result.get("unused_indexes", [])
-			if index["index_name"] not in ["parent", "creation"]
-		]
-
 		# convert all the float to int
 		for query in slow_queries:
 			for key, value in query.items():
 				if isinstance(value, float):
 					query[key] = int(value)
-		# sort the slow queries by `rows_examined`
-		result["slow_queries"] = sorted(slow_queries, key=lambda x: x["rows_examined"], reverse=True)
-		result["is_performance_schema_enabled"] = False
+		is_performance_schema_enabled = False
 		if database_server := frappe.db.get_value("Server", self.server, "database_server"):
-			result["is_performance_schema_enabled"] = frappe.db.get_value(
+			is_performance_schema_enabled = frappe.db.get_value(
 				"Database Server",
 				database_server,
 				"is_performance_schema_enabled",
 			)
+		result = None
+		if is_performance_schema_enabled:
+			with suppress(Exception):
+				# for larger table or if database has any locks, fetching perf report will be failed
+				result = agent.get_summarized_performance_report_of_database(self)
+				# remove `parent` & `creation` indexes from unused_indexes
+				result["unused_indexes"] = [
+					index
+					for index in result.get("unused_indexes", [])
+					if index["index_name"] not in ["parent", "creation"]
+				]
+
+		if not result:
+			result = {}
+			result["unused_indexes"] = []
+			result["redundant_indexes"] = []
+			result["top_10_time_consuming_queries"] = []
+			result["top_10_queries_with_full_table_scan"] = []
+
+		# sort the slow queries by `rows_examined`
+		result["slow_queries"] = sorted(slow_queries, key=lambda x: x["rows_examined"], reverse=True)
+		result["is_performance_schema_enabled"] = is_performance_schema_enabled
 		return result
 
 	@property

press/press/doctype/usage_record/usage_record.py

@@ -111,6 +111,7 @@ def link_unlinked_usage_records():
 			"invoice": ("is", "not set"),
 			"date": ("between", (fd, ld)),
 			"team": ("not in", free_teams),
+			"docstatus": 1,
 		},
 		pluck="name",
 		ignore_ifnull=True,