(crawler) Add timeout to wrap-up phase of WarcInputBuffer.

Since the crawl tasks are started in two phases, first when generating them in one loop, and then in a second loop that drains the task list; if the first loop contains a long-running crawl task that is triggered late, the rest of the crawl may halt until that task is finish.

Fixed the problem by draining and re-trying also in the first loop.

build.gradle

@@ -5,7 +5,7 @@ plugins {
 
     // This is a workaround for a bug in the Jib plugin that causes it to stall randomly
     // https://github.com/GoogleContainerTools/jib/issues/3347
-    id 'com.google.cloud.tools.jib' version '3.4.4' apply(false)
+    id 'com.google.cloud.tools.jib' version '3.4.5' apply(false)
 }
 
 group 'marginalia'
@@ -47,7 +47,7 @@ ext {
     dockerImageBase='container-registry.oracle.com/graalvm/jdk:24'
     dockerImageTag='latest'
     dockerImageRegistry='marginalia'
-    jibVersion = '3.4.4'
+    jibVersion = '3.4.5'
 }
 
 idea {

code/processes/crawling-process/java/nu/marginalia/crawl/CrawlerMain.java

@@ -264,17 +264,16 @@ public class CrawlerMain extends ProcessMainClass {
                 if (workLog.isJobFinished(crawlSpec.domain))
                     continue;
 
-                var task = new CrawlTask(
+                var task = new CrawlTask(crawlSpec, anchorTagsSource, outputDir, warcArchiver, domainStateDb, workLog);
-                        crawlSpec,
-                        anchorTagsSource,
-                        outputDir,
-                        warcArchiver,
-                        domainStateDb,
-                        workLog);
 
                 // Try to run immediately, to avoid unnecessarily keeping the entire work set in RAM
                 if (!trySubmitDeferredTask(task)) {
-                    // Otherwise add to the taskList for deferred execution
+
+                    // Drain the retry queue to the taskList, and try to submit any tasks that are in the retry queue
+                    retryQueue.drainTo(taskList);
+                    taskList.removeIf(this::trySubmitDeferredTask);
+
+                    // Then add this new task to the retry queue
                     taskList.add(task);
                 }
             }

code/processes/crawling-process/java/nu/marginalia/crawl/fetcher/warc/WarcInputBuffer.java

@@ -57,6 +57,7 @@ public abstract class WarcInputBuffer implements AutoCloseable {
             return new ErrorBuffer();
         }
 
+        Instant start = Instant.now();
         InputStream is = null;
         try {
             is = entity.getContent();
@@ -71,8 +72,25 @@ public abstract class WarcInputBuffer implements AutoCloseable {
             }
         }
         finally {
+            // We're required to consume the stream to avoid leaking connections,
+            // but we also don't want to get stuck on slow or malicious connections
+            // forever, so we set a time limit on this phase and call abort() if it's exceeded.
             try {
-                is.skip(Long.MAX_VALUE);
+                while (is != null) {
+                    // Consume some data
+                    if (is.skip(65536) == 0) {
+                        // Note that skip may return 0 if the stream is empty
+                        // or for other unspecified reasons, so we need to check
+                        // with read() as well to determine if the stream is done
+                        if (is.read() == -1)
+                            is = null;
+                    }
+                    // Check if the time limit has been exceeded
+                    else if (Duration.between(start, Instant.now()).compareTo(timeLimit) > 0) {
+                        request.abort();
+                        is = null;
+                    }
+                }
             }
             catch (IOException e) {
                 // Ignore the exception

code/services-application/api-service/build.gradle

@@ -3,7 +3,7 @@ plugins {
 
     id 'application'
     id 'jvm-test-suite'
-    id 'com.google.cloud.tools.jib' version '3.4.4'
+    id 'com.google.cloud.tools.jib' version '3.4.5'
 }
 
 java {

code/services-application/dating-service/build.gradle

@@ -3,7 +3,7 @@ plugins {
 
     id 'application'
     id 'jvm-test-suite'
-    id 'com.google.cloud.tools.jib' version '3.4.4'
+    id 'com.google.cloud.tools.jib' version '3.4.5'
 }
 
 application {

code/services-application/explorer-service/build.gradle

@@ -3,7 +3,7 @@ plugins {
 
     id 'application'
     id 'jvm-test-suite'
-    id 'com.google.cloud.tools.jib' version '3.4.4'
+    id 'com.google.cloud.tools.jib' version '3.4.5'
 }
 
 application {

code/services-application/search-service-legacy/build.gradle

@@ -5,7 +5,7 @@ plugins {
     id 'application'
     id 'jvm-test-suite'
 
-    id 'com.google.cloud.tools.jib' version '3.4.4'
+    id 'com.google.cloud.tools.jib' version '3.4.5'
 }
 
 application {

code/services-application/search-service/build.gradle

@@ -3,7 +3,7 @@ plugins {
     id 'application'
     id 'jvm-test-suite'
     id 'gg.jte.gradle' version '3.1.15'
-    id 'com.google.cloud.tools.jib' version '3.4.4'
+    id 'com.google.cloud.tools.jib' version '3.4.5'
 }
 
 application {

code/services-application/search-service/resources/jte/serp/error.jte

@@ -26,7 +26,7 @@
 
     <!-- Main content -->
     <main class="flex-1 p-4 max-w-2xl space-y-4">
-        <div class="border dark:border-gray-600 rounded bg-white text-black dark:bg-gray-800 dark:text-white text-m p-4">
+        <div class="border border-gray-300 dark:border-gray-600 rounded bg-white text-black dark:bg-gray-800 dark:text-white text-m p-4">
             <div class="flex space-x-3  place-items-baseline">
                 <i class="fa fa-circle-exclamation text-red-800"></i>
                 <div class="grow">${model.errorTitle()}</div>

code/services-application/search-service/resources/jte/serp/part/footerHowto.jte

@@ -80,10 +80,6 @@
             <tr><td>rank&gt;50</td><td>The ranking of the website is at least 50 in a span of 1 - 255</td></tr>
             <tr><td>rank&lt;50</td><td>The ranking of the website is at most 50 in a span of 1 - 255</td></tr>
 
-            <tr><td>count&gt;10</td><td> The search term must appear in at least 10 results form the domain</td></tr>
-            <tr><td>count&lt;10</td><td> The search term must appear in at most 10 results from the domain</td></tr>
-
-
             <tr><td>format:html5</td><td>Filter documents using the HTML5 standard. This is typically modern websites.</td></tr>
             <tr><td>format:xhtml</td><td>Filter documents using the XHTML standard</td></tr>
             <tr><td>format:html123</td><td>Filter documents using the HTML standards 1, 2, and 3. This is typically very old websites. </td></tr>

code/services-application/search-service/resources/jte/serp/part/searchform.jte

@@ -7,7 +7,7 @@
 
 <form class="flex-1 max-w-2xl" action="/search">
     <div class="flex">
-        @if (query.isBlank())
+        @if (query != null && query.isBlank())
         <%-- Add autofocus if the query is blank --%>
             <input type="text"
                    class="shadow-inner flex-1 dark:bg-black dark:text-gray-100 bg-gray-50 border dark:border-gray-600 border-gray-300 text-gray-900 text-sm rounded-sm block w-full p-2.5"

code/services-application/status-service/build.gradle

@@ -2,7 +2,7 @@ plugins {
     id 'java'
     id 'application'
     id 'jvm-test-suite'
-    id 'com.google.cloud.tools.jib' version '3.4.4'
+    id 'com.google.cloud.tools.jib' version '3.4.5'
 }
 
 java {

code/services-core/assistant-service/build.gradle

@@ -3,7 +3,7 @@ plugins {
 
     id 'application'
     id 'jvm-test-suite'
-    id 'com.google.cloud.tools.jib' version '3.4.4'
+    id 'com.google.cloud.tools.jib' version '3.4.5'
 }
 
 application {

code/services-core/control-service/build.gradle

@@ -2,7 +2,7 @@ plugins {
     id 'java'
     id 'application'
     id 'jvm-test-suite'
-    id 'com.google.cloud.tools.jib' version '3.4.4'
+    id 'com.google.cloud.tools.jib' version '3.4.5'
 }
 
 java {

code/services-core/executor-service/build.gradle

@@ -3,7 +3,7 @@ plugins {
 
     id 'application'
     id 'jvm-test-suite'
-    id 'com.google.cloud.tools.jib' version '3.4.4'
+    id 'com.google.cloud.tools.jib' version '3.4.5'
 }
 
 application {

code/services-core/index-service/build.gradle

@@ -3,7 +3,7 @@ plugins {
 
     id 'application'
     id 'jvm-test-suite'
-    id 'com.google.cloud.tools.jib' version '3.4.4'
+    id 'com.google.cloud.tools.jib' version '3.4.5'
 }
 
 application {

code/services-core/query-service/build.gradle

@@ -3,7 +3,7 @@ plugins {
 
     id 'application'
     id 'jvm-test-suite'
-    id 'com.google.cloud.tools.jib' version '3.4.4'
+    id 'com.google.cloud.tools.jib' version '3.4.5'
 }
 
 application {

code/tools/screenshot-capture-tool/build.gradle

@@ -3,7 +3,7 @@ plugins {
 
     id 'application'
     id 'jvm-test-suite'
-    id 'com.google.cloud.tools.jib' version '3.4.4'
+    id 'com.google.cloud.tools.jib' version '3.4.5'
 }
 
 java {

gradle/wrapper/gradle-wrapper.properties

@@ -1,5 +1,5 @@
 distributionBase=GRADLE_USER_HOME
 distributionPath=wrapper/dists
-distributionUrl=https\://services.gradle.org/distributions/gradle-8.10-bin.zip
+distributionUrl=https\://services.gradle.org/distributions/gradle-8.14-bin.zip
 zipStoreBase=GRADLE_USER_HOME
 zipStorePath=wrapper/dists